In the past few decades, two strategies for improving women?s political representation have become commonplace around the world: state feminism, which involves national policy machineries for improving gender equality, and gender quotas, which encompass national and party policies for increasing the number of female candidates for elected office. These two strategies seek to improve women?s representation in very different ways: state feminism aims to incorporate women's gender interests into public policy, thereby improving women's substantive representation, while gender quotas seek to improve women's access to electoral politics, thereby increasing women?s descriptive representation. Although recent work has analyzed the global diffusion of these strategies, as well as their adoption and implementation in individual countries, much less attention has focused on differences in their origins and outcomes in different world regions. Seeking to initiate a dialogue across research on the global North and South, we outline and compare trends in both policy areas in Western Europe and Latin America. We discover that in Western Europe, state feminism grew largely out of mobilization by women's movements and gender quotas were generally adopted voluntarily by political parties, while in Latin America, state feminism emerged from a combination of activism by women's movements, international organizations, and state actors and gender quotas were more often imposed through national legislation on all political parties. Probing these differences, we then utilize the comparison to assess the strengths and weaknesses of state feminism and gender quotas as strategies for improving women's descriptive and substantive representation. We conclude by arguing for more explicit cross-regional analysis of developments in gender and politics as a means for not only improving the study of women in politics, but also for devising more effective policies for incorporating women and women's concerns into the political arena. ..PAT.-Conference Proceeding [ABSTRACT FROM AUTHOR]