1: \begin{abstract}

2: When a granular system is tapped, its volume changes. Here, using a

3: well-defined macroscopic protocol, we prepare an ensemble of granular

4: systems and track the statistics of volume changes as a function of the

5: number of taps. This is in contrast to previous studies, which have focused

6: on single trajectories and assumed ergodicity.  We devise a new method to

7: assess the convergence properties of a sequence of ensemble volume

8: histograms and introduce a reasonable approximate version of an invariant

9: histogram. We then compare these invariant histograms with histograms

10: generated by sampling a long trajectory for one system and observe

11: nonergodicity, which we quantify. Finally, we use the overlapping histogram

12: method to assess potential compatibility with Edwards' canonical

13: assumption. Our histograms are incompatible with this assumption.

14: \end{abstract}

15: