1: \begin{abstract}

2: This paper presents an alternative approach for the computation of trajectory segments on slow manifolds of saddle type. This approach is based on iterative methods rather than collocation-type methods. Compared to collocation methods, that require mesh refinements to ensure uniform convergence with respect to $\epsilon$, appropriate estimates are directly attainable using the method of this paper. The method is applied to several examples including: A model for a pair of neurons coupled by reciprocal inhibition with two slow and two fast variables and to the computation of homoclinic connections in the FitzHugh-Nagumo system. %\ed{The method is also applied to an example where Fenichel's theory does not apply but where there exists an almost invariant slow manifold. }

3: % In this paper we present some results on the numerical implementation of some methods developed by the authors elsewhere for the computation of slow manifolds. A particular focus will be on demonstrating how the approximation of fibers can be used to compute fairly accurate approximations of solutions initially approaching and finally excaping canard solutions. 

4: \end{abstract}

5: