Every knot in the 3-sphere can be realized as a cross-section of some unknotted surface in the 4-sphere. For a given knot, the least genus of any of such surface is defined to be its double slice genus. Obviously twice the slice genus of a knot is a lower bound for its double slice genus. One really basic question is whether the double slice genus can be arbitrarily large compared to twice the slice genus. However, this was not answered due to the lack of lower bounds for the double slice genus. In this talk I will introduce a lower bound that can be used to answer this question.