This paper applies option pricing analysis to the problem of valuing the abandonment option of an investment proposal. The assumption is made that the abandonment option is exercisable at only one point in time in the future and that the project's vatue-in-use and its abandonment value are lognormally distributed. The model is employed to measure how the uniqueness of the project asset, as measured by the correlation between these two lognormal random variables, affects the value of the abandonment option. It is shown that the more unique the asset, or the higher the correlation, the lower is the value of the abandonment option. The model is also employed to examine the impact of increased uncertainty in these two random variables on the value of the abandonment option. The relationships are shown to be nonmonotonic. However, beyond critical thresholds, increased uncertainty in either one of the two variables enhances the value of the abandonment option.