The climatic index of dune mobility developed by Lancaster (1988) has been applied to a variety of different environments. The index is, however, untested and unverified. We tested the index by comparison of values of the dune mobility index calculated from climate data with rates of sand transport measured at three stations in Arizona and New Mexico over the period 1985 to 1997. Our results show that changes in measured rates of sand transport closely parallel temporal changes in the dune mobility index. The mobility index is, however, a relatively poor predictor of the magnitude of actual sand transport on a year-to-year basis. This discrepancy is probably due to the fact that sand transport rates at these sites are strongly influenced by vegetation cover, the state of which may lag changes in annual precipitation. There is, however, a good relation between the mean annual mobility index and mean annual rates of sand transport. This indicates that the dune mobility index is a valid predictor of the long-term state of the aeolian system and can be used confidently for the purposes for which it was originally intended. Copyright (C) 2000 John Wiley and Sons, Ltd.