The UK accounting standard SSAP 13 Accounting for Research and Development was revised in 1989 making the disclosure of Research and Development (R&D) expenditure mandatory for the first time. The change provided the impetus for the research undertaken in this thesis which comprises three pieces of empirical work. A survey of the reporting of R&D information in UK annual reports was conducted which covered the period 1985-1990. This was more comprehensive and detailed than any of the available survey material. The survey revealed that there was a significant number of companies failing to comply with SSAP 13 (revised), either by failing to disclose R&D expenditure or by continuing to disclose it in the directors' report rather than as part of the audited financial statements. The revision of SSAP 13, a revision which only changed disclosure requirements, appeared to have caused some firms to change their accounting policy. A limited dependent variable model was then used to investigate various hypotheses concerning the decision to disclose R&D expenditure over the period 1981-1990 when disclosure was voluntary. The main contribution of this part of the thesis was a logit model of the decision to disclose R&D expenditure which performed well in terms of predictive accuracy, goodness of fit, stability and lack of problems of heteroskedasticity. The model supported some hypotheses which related to discretionary disclosure in general and others which incorporated the special nature and commercial sensitivity of R&D expenditure information. Finally, an event study of the market's response to the announcement of R&D information was conducted. The revision of SSAP 13 meant that for the first time there was a large enough set of observations to make such a study possible. The market will only respond to the extent that it is surprised. This study investigated the association between cummulative abnormal returns (CARs) and various measures of R&D expenditure surprise in regressions which took careful account of the information already available to the market and the other information released on the event date. The main results were that the market responded positively to an unexpected increase in R&D expenditure but that this positive response was more in evidence pre-1989 when disclosure was voluntary.