Some basic characteristics of ridge axis topography are related to spreading rate and distance from neighboring transform faults. For example, the presence of an axial depression coincides in most cases with slow spreading rates, and the overall depth of the ridge axis increases toward ridge-transform intersections (RTIs). On the other hand, it is also well known that the relief and width of the axial valley on, say, the Mid-Atlantic Ridge (MAR) vary along strike in an unpredictable manner. The purpose of the present study is to quantify how much of the observed variation in the first-order topography at the axis is related to changes in other parameters, such as spreading rate and distance from RTIs. To carry out this test, the zero-age depth and the relief and width of the axial valley have been estimated on 46 profiles that cross the axis of the MAR between the equator and 50¿N (full spreading rates 22--36 km/Ma). Zero-age depth is here defined to be the depth at age zero of the best fit thermal subsidence trend. Axial valley relief and width have been measured with respect to the ridge flank by the least squares fit of a Gaussian bell.

The measured axial valley relief varies between 600 and 2100 m (average ~1300 m), while the valley width varies between 16 and 62 km (average ~35 km). The correlation between zero-age depth, axial valley relief and width, latitude of axial crossing, spreading rate, distance from nearest RTI, and offset on the nearby transforms has been investigated using linear regression techniques. The main results of the present study are that (1) zero-age depth significantly correlates with latitude of crossing, distance from nearest RTI, and offset on the nearby transforms; and (2) the variation in axial valley relief and width is essentially uncorrelated with spreading rate, zero-age depth, distance from nearest RTI, and offset on nearby transforms. The preferred explanation for the observed spatial variation in axial valley geometry is that it reflects a temporal variation. In fact, if the rough abyssal hill topography typical of the MAR flanks is created within the axial valley, the shape of the axial valley cannot be steady state (although the existance of an axial valley may be a steady state phenomenon). This hypothesis is supported by the observation that the variability in axial valley relief is similar to the overall amplitude of abyssal hill topography, measured as the residual on the thermal subsidence trend. ¿ American Geophysical Union 1990