The paper is devoted to the estimation of limited dependent variable models with dependent observations. This topic has received relatively little attention due to the computational complexity of the maximum likelihood estimator. A computationally attractive and relatively efficient alternative utilizing orthogonality conditions is developed. The resulting generalized conditional moment estimators can be applied with a known or an unknown disturbance covariance matrix. Although the paper considers only the probit model, the approach may be generalized to other limited dependent models.