Experimental evidence for very long term processes

Well, if it is so easy for new complex biological features to evolve by RV plus NS because they are countless, then it should be trivially easy to rise to Bechly’s challenge of the species pairs. But is it?