This paper reports the distribution of agreements over time in four bargaining experiments. There is a striking concentration of agreements reached in the very last seconds before the deadline. This deadline effect appears to be quite robust, in that the distribution of agreements over time appears to be much less sensitive to the experimental manipulations than is the distribution of the terms of agreement. Each of these experiments also exhibited a substantial frequency of disagreement. Since last-minute agreements are widely believed to occur frequently in naturally occurring negotiations,' it may be helpful to