Time and space are continuous challengers for the researcher, particularly the cross-nationally oriented one, time being the more uneasy companion in the present context. There could be no greater contrast between cross-national research on coalition formation and government stability after World War II and the virtual neglect of this topic for the interwar period. The efforts of De Swaan' and Dodd2 are about the only exceptions to this verdict. De Swaan, however, deals only with coalition formation. Dodd covers twelve European countries in the interwar period. He is concerned with cabinet coalitional status and cabinet durability, defined in terms of lasting coalitions, and less with the parliamentary correlates of government duration assessed in this article. The present paper will not be a remedy to this poor state of research. It will, however, for the first time provide empirical findings on a more systematic, although still highly limited, base. Thus far data on parliamentary correlates of coalition formation and government duration have been collected for Germany, Austria, France, the United Kingdom, Belgium, and the Netherlands. Only part of these data will be analyzed here.3 Also, further checks on the quality of some of these data seem to be indicated, even though it is unlikely that the structure of results will change considerably. What might change, however, is the overall evaluation of post-World War II coalition theories when applied to the interwar period once the present sample is broadened into a study of all European democracies. Present findings provide various hints for such a reorientation. The selection of countries in this instance was dictated by the notion of dense area sampling, trying to follow the logic of most similar systems designs and thus allowing for a variety of quantitative and qualitative comparisons.4 Consequently, tests of significance are not applied. The study of government duration is important for at least four major reasons. First, in parliamentary democracies (and elsewhere) governments are among the prime actors, if not the predominant one that needs to be taken into consideration when assessing macropolitical (research) problems. Second, polities obviously differ in the economic and output of their governments. Third, they also differ in how people respond to both these outputs and their governments. All three factors -governments, outputs, and responses-have also been summarized in the notion of political performance (political outcomes, policy effectiveness). Fourth, the survival of polities as such cannot be dissociated from the study of governments, even though government failure in liberal democracies is only a necessary, not sufficient cause for polity collapse. Government output as well as responses of citizens have been extensively studied on a