The availa. bility of a variety of sources of evidence on phylogenetic relationн ships has focused attention on a fundamental question: Should different kinds of data bearing on a given phylogenetic problem be analyzed separately or combined and analyzed simultaneously? Our aim is to review arguments for and against separate and combined analyses and to provide a general concepн tual framework within which to explore the basic but sometimes subtle issues associated with the problem. We hope to show that both kinds of analyses are useful in f! stimating phylogenetic relationships, but that the standard forms of these altematives by no means exhaust the possible solutions to the problem. Some additional clarification of our goals is in order, especially to say what we do not intend to provide. First, although we need to refer to particular examples in order to clarify arguments, we do not provide a thorough …