Recent high precision experimental data for heavy-ion fusion reactions at subbarrier energies systematically show that a surprisingly large surface diffuseness parameter for a Woods-Saxon potential is required in order to fit the data. We point out that experimental data for quasi-elastic scattering at backward angles also favor a similar large value of surface diffuseness parameter. Consequently, a double folding approach fails to reproduce the experimental excitation function of quasielastic scattering for the 16 O + 154 Sm system at energies around the Coulomb barrier. We also show that the deviation of the ratio of the quasielastic to the Rutherford cross sections from unity at deep subbarrier energies offers an unambiguous way to determine the value of the surface diffuseness parameter in the nucleus-nucleus potential.PACS numbers: 25.70. Bc,25.70.Jj,24.10.Eq,27.70.+q The nucleus-nucleus potential is the primary ingredient in nuclear reaction calculations. Its nuclear part has often been parametrized as a Woods-Saxon form [1]. Elastic and inelastic scattering are sensitive mainly to the surface region of the nuclear potential, where the WoodsSaxon parametrization has a simple exponential form. This fact has been exploited to study the surface property of nuclear potential. Usually, the best fit to experimental data for scattering is obtained with a diffuseness of around 0.63 fm [1,2,3,4,5]. This value is consistent with a double folding potential [6,7], and seems to be well accepted [1,8].In marked contrast, recent high precision experimental data for heavy-ion fusion reactions at energies around the Coulomb barrier suggest that a much larger value of diffuseness, ranging from 0.75 to 1.5 fm, is required to fit the data [6,7,9,10,11,12] (See Ref.[13] for a detailed systematic study). The Woods-Saxon potential which fits elastic scattering overestimates fusion cross sections at energies both above and below the Coulomb barrier, having an inconsistent energy dependence to the experimental fusion excitation function. When the height of the Coulomb barrier is fixed, the larger diffuseness parameter leads to the smaller barrier position and the smaller barrier curvature (thus the larger tunneling region). The main effect on the fusion cross sections comes from the barrier position and the tunneling width of the barrier at energies above and below the Coulomb barrier, respectively. A large diffuseness parameter appears to be desirable in both these aspects [6]. The reason for the large discrepancies in diffuseness parameters extracted from scattering and from fusion analyses, however, has not yet been understood.The purpose of this paper is to discuss the dependence of quasielastic excitation function at a large scattering angle on the surface diffuseness parameter in a nucleusnucleus potential. The quasielastic cross section is defined as the sum of the cross sections of elastic, inelastic, and transfer reactions. Its excitation function at backward angles provides complementary information to the fusion proc...