Reference
Briggs, W.M. 2008. On the changes in the number and intensity of North Atlantic tropical cyclones. Journal of Climate 21: 1387-1402.
What was done
"Bayesian statistical models," in the words of the author, "were developed for the number of tropical cyclones, the rate at which these cyclones became hurricanes, and the rate at which hurricanes became category 4+ storms in the North Atlantic using data from 1966 to 2006 and from 1975 to 2006."
What was learned
Briggs finds that "to conclude that there has been an increase in the number of tropical cyclones in the North Atlantic basin depends on the starting date used for the analysis," because "the number of storms circa 1966 were in a relatively high period, while those circa 1975 were in a relatively low period; and, of course, circa 2005 is a relatively high period." Consequently, he notes that "starting from 1966 does not give evidence of a linear increase, but starting from 1975 does," perhaps because of "some underlying cyclicity in the data that coincidentally bottomed out around 1975."
Briggs also finds "no evidence that the distributional mean of individual storm intensity, measured by storm days, track length, or individual storm power dissipation index, has changed (increased or decreased) through time." What has increased, in his words, is "the distributional variance of individual storm intensity," which suggests that "there should be more storms with higher maximum wind speeds, but there should also be more storms with lower maximum wind speeds."
What it means
Based on the results of this study, it would appear that the global warming of the past four decades has not had a major impact on the number of tropical cyclones in the North Atlantic basin, nor has it affected their overall mean intensity.