that supports the Alabamian application of gap for saddle

I was just concerned with the notion that water would flow through a saddle; given that a saddle has negative curvature (I defined it, Mr. M., it ain't no straight line) the water would have to flow uphill at some point for it to be a water gap.