Comment on "Comparison of approaches to classical signature change"

This is a comment on a reply (gr-qc/9601040) to a comment (gr-qc/9606045) on a paper of Hellaby & Dray (gr-qc/9404001), repeating the identification of an important mistake which is still being denied by the authors: their proposed solutions do not satisfy the Einstein- Klein-Gordon equations at a change of signature. Substitution of the proposed solutions into the Einstein-Klein-Gordon equations in unit normal coordinates yields Dirac delta terms describing source layers at the junction. Hellaby & Dray's criticisms of this straightforward calculation are absurd: it does not involve "imaginary time", it does not involve a "modified form" of the field equations, and it is "purely classical". Moreover, Hellaby & Dray's latest attempt to lose the delta terms is mathematically invalid, involving division by zero and products of distributions, hinging on an identity whose incorrectness may be checked by substitution.