The case of logic-based agents that need to communicate in order to coordinate is considered. By logic-based it is meant that these agents represent their beliefs about the world in a database consisting of logical propositions, and use logical deduction to construct sequences of actions to be taken in the world. As an agent receives a message and adds propositions to its database, it may be able to form a plan that is coordinated with the actions of the agent that sent the message. The point of communication for these agents, then, is to bring about coordination through the passing of propositions. Several communication strategies are outlined. Some of these strategies are proved to be convergent (i. e. , the agents, using the strategy, will converge to an agreed-upon plan), while others are not. An analysis is also made of techniques for resolving disagreement among the agents' databases, and the potential advantages of passing false information.