Fast and accurate analysis of flourescence in-situ hybridization (FISH) images will depend mainly upon two components: a classifier to discriminate between artifacts and valid signal data, and well discriminating features to represent the signals. Our previous work has focused on the first component. To investigate the second component, we evaluate candidate feature sets by illustrating the probability density functions and scatter plots for the features. This analysis provides insight into dependencies between features, indicates the relative importancce of members of a feature set, and helps in identifying sources of potential classification errors. The analysis recommends several intensity and hue-based features for representing FISH signals. The recommendation is confirmed by the probability of misclassification using a two-layer neural network (NN), and also by a feature selection technique making use of a class separability criterion. Represented by these intensity and hue-based features, 90% of valid signals and artifacts are corrently classified using the NN.