The first definitive strepsipteran is reported from the Cretaceous, named Cretostylops engeli, n.gen., n.sp., which is an adult male in amber from the mid-Cretaceous (approximately Cenomanian) of northern Myanmar (Burma). A triungulin from the Late Cretaceous (Campanian, c. 80 myo) of Manitoba, Canada is possibly a strepsipteran. The triungulin is described in detail but its morphology does not conform to any known clade of Recent strepsipterans. Other Cretaceous triungula reported here are in Burmese amber and are probably of the family Rhipiphoridae (Coleoptera), and bizarre (possibly coleopteran) triungula in mid-Cretaceous (Turonian, c. 90 myo) amber from New Jersey, USA. Phylogenetic analysis confirms the primitive position of Cretostylops among families of Strepsiptera, but it is not as primitive as Protoxenos in Eocene Baltic amber. Protoxenos and Cretostylops are still too highly modified to address the controversial relationships of Strepsiptera among insect orders, but the generalized structure of the mandible is inconsistent with the hypothesis that this order is the sister group to Diptera or closely related to Mecopterida. Phylogeny of living and Recent Strepsiptera suggests an origin of the order in the Early Cretaceous or Late Jurassic, which is also inconsistent with this order being a sister group to the much older Diptera.