Reply by Author to A. E. Bryson Jr.
Cecil D. Bailey · AIAA Journal · 1976
where the integrals are line integrals in the qlt q1 phase space; thus the work done is path -dependent. The solution to the second example appears to be correct, indicating that Bailey again used virtual work, not work, in determing the solution. We agree with Bailey that there is some confusion on this subject in the recent literature. For example, Goldstein has an incorrect derivation of Lagrange's equations for nonconservative systems using a principle similar to that of Ref . 1 and Lanczos apparently chooses not even to discuss the issue in his treatise. However, Whittaker in 1904 gave a complete discussion (see Ch. IX of Ref. 2).