This paper examines how private information affects trading volume, the information content of trading volume data, and if there are any relations between trading volume and price changes which can be explained by informational differences. We develop a model with two trading periods in which asymmetrically informed agents learn the trading volume of the first round of trading prior to trading in the second round. We also develop a general method for computing equilibria in such a model assuming only smooth, concave utility functions, and asset return distributions and signals with smooth densities. The model prices (i) a positive relation between trading volume and the absolute value of price changes; (ii) a positive relation between trading volume and subsequent stock price volatility; and (iii) that positive price moments on high trading volume lead, on average, to positive price movements.