The object of this paper is to study estimates of −q W p (µ+ ν, µ) for small > 0. Here W p is the Wasserstein metric on positive measures, p > 1, µ is a probability measure and ν a signed, neutral measure ( dν = 0). In [W1] we proved uniform (in ) estimates for q = 1 provided φdν can be controlled in terms of |∇φ| p/(p−1) dµ, for any smooth function φ.In this paper we extend the results to the case where such a control fails. This is the case where if, e.g. µ has a disconnected support, or if the dimension of µ , d (to be defined) is larger or equal p/(p − 1).In the latter case we get such an estimate provided 1/p + 1/d = 1 for q = min(1, 1/p + 1/d). If 1/p + 1/d = 1 we get a log-Lipschitz estimate.As an application we obtain Hölder estimates in W p for curves of probability measures which are absolutely continuous in the total variation norm .In case the support of µ is disconnected (corresponding to d = ∞) we obtain sharp estimates for q = 1/p ("optimal teleportation"):where ν µ is expressed in terms of optimal transport on a metric graph, determined only by the relative distances between the connected components of the support of µ, and the weights of the measure ν in each connected component of this support.