Did insect pollination cause increased seed plant diversity?

The dominant paradigm for the disproportionate number of flowering plants is the unique coevolution that they underwent with pollinating insects. The theory underlying this biotic pollination hypothesis contradicts more generally accepted evolutionary theory. Furthermore, various lines of empirical evidence falsify the biotic pollination hypothesis: (a) several lineages of plants were insect pollinated – angiosperms, Gnetales, Bennettitales, Cheiro-lepidiaceae, Medullosales, and Cycadales - yet only the first four were ever diverse or underwent radiations; (b) the predicted rise in insect diversity, which was coupled with angiosperm radiation, does not appear in the fossil record; (c) the family Poaceae (grasses) are wind pollinated, yet are exceptionally diverse and species-rich; and (d) the family Formicidae (ants) were not pollinators, yet are exceptionally species-rich and ecologically dominant. I enumerate many alternate (and seldom investigated) hypotheses for these patterns of seed plant diversity, keeping in mind that although I show that insect pollination was neither a necessary nor sufficient condition for large numbers of species, it may have played a substantial role in both plant and insect speciation. Alternatively, existing theory regarding the biotic pollination hypothesis can be refined in an attempt to eliminate the aforementioned empirical anomalies and theoretical inconsistencies.