Answers are grounded in this paper and cite their sources.
The author deals with the mutual inequivalence of g-variation and P-variation of a given action S, the components of linear connection Gamma m kl being required to be symmetric. Under g-variation S is required to be stationary with respect to variations of the metric tensor g ij , the Gamma m kl being taken to be Christoffel symbols from the outset, whereas under P-variations the g ij and Gamma m kl are initially regarded as mutually independent of S is required to be stationary with respect to independent variations of these quantities. The discussion is illustrated at length by examples in which the Lagrangian of S is one or another of a set of homogeneous or inhomogeneous quadratic invariants of the Riemann tensor.
No takes yet. Share an insight, caveat, or question.
H. A. Buchdahl (1979) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: