Classification of otoliths to a shape (either comb, mixed, or hatchet) using a linear discriminant analysis (LDA), compared to shapes identified visually by an experienced age reader (Visual ID).

The sample size is the number of samples identified as each shape using visual ID. The number and % under LDA are the sample size and % of the total sample that were identified as each shape using a leave-one-out jack-knife cross-validation. Cells in grey indicate that the visual ID and the LDA were in agreement.