An efficient, fully-distributed protocol for integrated voice/data traffic in a local-area, random-access broadcast network is described in which time frames are dynamically partitioned between voice and data according to the traffic load. The protocol introduces a movable voice-data boundary to framed TDMA/CSMA and eliminates the requirement of system-wide synchronized clocks. The movable boundary is a major advantage in any system where fluctuations in voice and data loads are expected because assignment of idle capacity from one traffic class to the other increases the utilization of the channel. The protoco; provides collision-free virtual circuits for voice and periods of non-persistent CSMA/CD for data traffic and call establishment, and can support multi-party calls as well as two-way conversations. The protocol supports variable-size voice packets with very low overhead, and variable-size data packets that may be much longer than voice packets. This is of significant practical advantage over previous work, which has required fixed-size voice and/or data packets, or voice packets with significantly higher overhead, or severe restrictions on data-packet length. The behavior of framed TDMA/CSMA in general, and movable-boundary framed TDMA/CSMA in particular, is investigated via simulation and analysis. A method of dynamically controlling the movable boundary to balance the voice and data traffic is also proposed; this provides for maximum voice utilization consistent with user-defined constraints on data delay.