Hi, I am using the function lda() from MASS for finding reduced-dimensional representations of a datset. In reading various texts to compare LDA with Fisher's LDA approach (including Ripley's Modern Applied Statistics with S-Plus), it is still unclear to me whether or not they produce the same classification results, how they are related, and which is being performed by function lda(). One interpretation I have is that Fisher's LDA rule for classification is the same as the probabilistic Bayes approach to LDA (assumes classes are distributed Gaussian with equal covariance) when the prior probabilities are equal. If this is the case, then function lda() would be the same as Fisher's method when priors are equal. Is this correct? Thanks, Chris Harle H. John Heinz III School of Public Policy & Management Carnegie Mellon University