In digital electronics, transmission gates are transistor-level implementations of selectors, just as multiplexers were implemented at the gate-level.
The motivation behind this is that a 2-to-1 mux will take up 14 transistors. If we instead go to the transistor-level, we can implement the same functionality with 6 transistors. The design of a circuit like this aims to also prevent weak ones.
Then we can implement a 2-to-1 mux:
