The purpose of these experiments J was to determine the shape of the interaction curve for axial buckling vs torsional buckling of conical shells. In this type of combined loading situation a large number of specimens is required to explore the numerous parameters involved. This problem was overcome through the use of Mylar conical shells that could withstand repeated buckling tests without permanent damage. The data collected from these tests were compared with the various theories of failure for axial and torsional buckling loads acting alone. The combined loading curves for conical shells were shown to be similar to those that have been proposed for cylindrical shells. The amount of scatter in the data was within the bounds suggested by a correction factor that considers the initial imperfections in the shell. This correction factor proved to be a lower bound for all data accumulated in these tests.