1: \begin{abstract}

2: 

3: 

4: 	The Dynamical Cluster Approximation (DCA) is modified to include

5: disorder.    The DCA incorporates non-local corrections to local 

6: approximations such as the Coherent Potential Approximation (CPA) by 

7: mapping the lattice problem with disorder, and in the thermodynamic limit, 

8: to a self-consistently embedded finite-sized cluster problem.  It satisfies 

9: all of the characteristics of a successful cluster approximation.  It is 

10: causal, preserves the point-group and translational symmetry of the 

11: original lattice, recovers the CPA when the cluster size equals one, and 

12: becomes exact as $N_c\to\infty$.  We use the DCA to study the Anderson 

13: model with binary diagonal disorder.  It restores sharp features and band 

14: tailing in the density of states which reflect correlations in the local 

15: environment of each  site.  While the DCA does not describe the localization

16: transition,  it does describe precursor effects of localization.

17: 

18: \end{abstract}

19: