A SEEMING paradox has puzzled investigators of the crustal structure of the Gulf of Mexico since Ewing et al.1 calculated that a unit area of the rather thick crust in the gulf contains less mass than does a combination of the crust and enough of the upper mantle to make a comparable thickness in the Atlantic Ocean. They also noted that the free-air gravity of the gulf is essentially normal and fails by a large factor to be low enough to reflect the mass difference that they calculated. We propose a solution to this problem.