Abstract
On the basis of an observation from the binomial distribution with known probability of success p, we construct two classes of admissible estimators of the parameter n∈ { 1,2,… } when the loss function is the squared error. The estimators are either proper Bayes or limits of Bayes estimators.