In this paper we follow a different approach by taking a first step towards an option valuation model which does not explicitly make use of unobservable State variables. Instead of using a stochastic variance variable directly, we assume that the variance of stock returns is determined by the trading activity in the stock or the options market, respectively. As we will see, this is consistent with many empirical studies which report a positive relationship between the volume and volatility of individual securities. To our knowledge, this is the first time that an option pricing model uses trading volume in order to model the stochastic nature of the stock return variance. The major focus of our work is to combine recent theoretical work on option pricing models with results from the Statistical literature on volume, volatility and stock returns.