The nature of the relationship between trading volume and volatility series has been widely studied

from a short-run or a long-run perspective, but overall the literature provides mixed results. We investigate

this issue for the thirty components of the Dow Jones stock market index in the light of a recent

general p-component model who can be related to the theoretical financial literature on the transmission

mechanisms of the information flow. Detecting power law in coherency at frequencies beyond zero, our

analysis shows that trading volume and volatility are linked through a persistent common factor dwarfed

by more persistent idiosyncratic components in the case of more than half of the firms under analysis.

In contrast to cointegration theory, this phenomenon is compatible with both the mixture of distributions

hypothesis and the sequential arrival of information hypothesis. A subsequent non-parametric phase

spectrum analysis reveals that in all cases the former hypothesis is retained.