This paper examines how interpersonal social networks help explain the voting behavior of men and women. We argue that the gender gap in voting is influenced by the partisan and gender composition of networks, rather than just the latter. Building on this foundation, we explain how gendering in network construction and impact helps create a cleavage between men and women even under conditions that are often close to "random mixing." Analysis of the 2000 American National Election Study shows the voting gap is related to men excluding women from political networks, men being less exposed to females who support Democrats, and men being more strongly influence by women who support Republicans. The principal conclusion of the paper is that the role of social networks in explaining gendered voting is a function of combined partisan and gender segregation, principally by men.