We show that the positive volatility-volume relation documented by num
erous researchers actually reflects the positive relation between vola
tility and the number of transactions. Thus, it is the occurrence of t
ransactions per se, and not their size, that generates volatility; tra
de size has no information beyond that contained in the frequency of t
ransactions. Our results suggest that theoretical research needs to en
tertain scenarios in which (i) both the frequency and size of trades a
re endogeneously determined, yet (ii) the size of trades has no inform
ation content beyond that contained in the number of transactions.