In detail, the **Stone space** of a Boolean algebra *A* is the set of all 2-valued homomorphisms on *A*, with the topology of pointwise convergence of nets of such homomorphisms. Every Boolean algebra is isomorphic to the algebra of clopen (i.e., simultaneously closed and open) subsets of its Stone space. The isomorphism maps any element *a* of *A* to the set of homomorphisms that map *a* to 1.

Every totally disconnected compact Hausdorff space is homeomorphic to the space of 2-valued homomorphisms on the Boolean algebra of all of its clopen subsets. The homeomorphism maps each point *x* to the 2-valued homomorphism φ given by φ(*S*) = 1 or 0 according as *x* ∈ *S* or *x* not ∈ *S*. (Perhaps this is one of the few occasions for such rapid-fire mulitple repetition of the two distinct words *homomorphism* and *homeomorphism* in one breath. Let us therefore warn the reader not to confuse them with each other.)

Homomorphisms from a Boolean algebra *A* to a Boolean algebra *B* correspond in a natural way to continuous functions from the Stone space of *B* into the Stone space of *A*. In other words, this duality is a contravariant functor.