This thesis examines disclosure of 250 profit forecasts in 701 UK takeover bids in the period 1988 to 1992 against five research issues: - Factors influencing disclosure of forecasts - Influence of prevailing market expectations - Effect of disclosure of forecasts on the outcome of bids - Factors influencing disclosure content in forecasts - Whether forecasts disclosed convey good news Logit analysis and negative binomial regression are the two primary statistical techniques used to analyse the results. Results show the domination of the takeover-context of the research. Two variables accounted for almost all the influence on disclosure of forecasts for both bidders and targets: bid horizon and type of bid. Probability of disclosure of a forecast is greater the shorter the bid horizon and during contested bids. In addition to bid horizon and type of bid, for bidders, year, value of bid and purchase consideration were significant, and for targets value of bid and industry were significant in one of the two models estimated. Evidence supporting the hypothesis that forecast disclosure is more likely when market expectations are out of line with actual results is provided. There is some evidence that forecasts by targets affect the outcome of bids, but there is no such evidence for bidders. Takeover-context variables and forecast-related variables were most relevant in determining disclosures in forecasts. Disclosure content in forecasts was significantly greater during contested bids, in voluntary forecasts and in longer period forecasts. Significantly more assumptions were disclosed by target forecasters and in longer horizon forecasts. Evidence shows a tendency to disclose good news, with some disclosure of bad news. Good news forecasts are more likely during contested bids. Targets are more likely to disclose bad news forecasts, but when bidders disclose bad news it tends to be worse on average than targets' bad news.