Abstract interpretation techniques are used to derive a control-flow analysis for a simple higher-order functional language. The analysis approximates the interprocedural control-flow of both function calls and returns in the presence of first-class functions and tail-call optimization. The analysis is systematically derived by abstract interpretation of a stack-based abstract machine using a series of Galois connections. We prove that the analysis is equivalent to an analysis obtained by first transforming the program into continuation-passing style and then performing control flow analysis of the transfored program. We then show how the analysis induces an equivalent constraint-based formulation, thereby providing a rational reconstruction of a constraint-based CFA from abstract interpretation principles.