Christiano and Fitzgerald (2003) found a significant, positive correlation between M2 money growth and CPI inflation in all examined frequency bands for the U.S. prior to 1961. However, for post-1960 data, they found a positive correlation only in the frequency band corresponding to cycles of 20-40 years. Using their filter, we verify this result and extend the pre-1961 sample to include the monetary base and inflation calculated from the GDP deflator. In addition, we extend their post-1960 analysis to include growth in the monetary base, M1, and M3. A strongly positive correlation between post-1960 money growth and inflation exits only for the broad money aggregates and within the 20-40 year frequency band.