Abstract
We present two different approaches to evaluating the probability of finding in a particular state two neighboring lattice sites on a one-dimensional array of dumbbells. The results we find are exact and the same with either the first or the second procedure.