Consider the density of magnetic field lines (magnetic flux), assuming the opposite poles are at apex and base, that the lines are conserved and that the majority emanate from the poles.
Forgive my ignorance .. I'm FAR from being a physicist ..
"that the majority emanate from the poles" .. this would imply the flux density would be greater at the apex then the base ..
if one envisions a parrallel alignment of magnetic domains within the conical shape .. it would seem reasonable to assume more flux lines emininate from the "sides". If this were the case the flux density would be less at the apex then the base ..