Further study is done of Holloway and Gargett's (1987) suggestion that it is possible to distinguish salt-fingering microstructure signals from those induced by turbulence by using the kurtosis (the relative flatness) of the scalar probability distribution. Support for their result comes from an examination of towed measurements of conductivity gradient fluctuations from turbulent and salt finger patches in the Sargasso Sea; however, when data are examined from presumably salt finger interfaces in the thermohaline staircase east of Barbados (the Caribbean Sheets and Layers Transects (C-SALT) experimental region), data segments are found having both the suppressed kurtosis values expected of salt fingering and the high values associated with turbulence. It is concluded that either turbulence or some other chaotic process is competing with salt fingering in the C-SALT data or that the kurtosis discriminant is not as useful as Holloway and Gargett suggest. ¿ American Geophysical Union 1988