Bridges are used to divide larger networks into smaller sections. They do this by sitting between two physical network segments and managing the flow of data between the two. By looking at the MAC address of the devices connected to each segment, bridges can elect to forward the data (if they believe that the destination address is on another interface), or block it from crossing (if they can verify that it is on the interface from which it came). Figure 4 shows how a bridge can be used to segregate a network.
When bridges were introduced, the MAC addresses of the devices on the connected networks had to be entered manually, a time-consuming process that had plenty of opportunity for error. Today, almost all bridges can build a list of the MAC addresses on an interface by watching the traffic on the network. Such devices are called learning bridges because of this functionality.