Because so many species exist in such diversity, the higher classification of weevils is in a state of flux. They are generally divided into two major divisions, the Orthoceri or primitive weevils, and the Gonatoceri or true weevils (
Curculionidae).
E. C. Zimmerman proposed a third division, the Heteromorphi, for several intermediate forms. Primitive weevils are distinguished by having straight antennae, while true weevils have elbowed (geniculate) antennae. The elbow occurs at the end of the
scape (first antennal segment) in true weevils, and the scape is usually much longer than the other antennal segments. Some exceptions occur, such as
Nanophyini, primitive weevils with long scapes and geniculate antennae, while among the true weevils,
Gonipterinae and
Ramphus have short scapes and little or no "elbow". A 1995 classification system to family level was provided by Kuschel, with updates from Marvaldi
et al. in 2002, and was achieved using
phylogenetic analyses. The accepted families were the primitive weevils,
Anthribidae,
Attelabidae,
Belidae,
Brentidae,
Caridae, and
Nemonychidae, and the true weevils
Curculionidae. Most other weevil families were demoted to subfamilies or tribes. Further work resulted in the elevation of
Cimberididae to family from placement as a subfamily of Nemonychidae in 2017 and the recognition of the
Cretaceous age family
Mesophyletidae in 2018 from
Burmese amber. The oldest weevils date to the Middle-Late Jurassic boundary, found in the
Karabastau Formation of Kazakhstan, the
Shar-Teg locality of Mongolia, the
Daohugou locality in Inner Mongolia, China, and the
Talbragar site in Australia. The extinct family
Obrieniidae, with species dating from the
Ladinian stage of the
Triassic through to tentatively the
Oxfordian, have sometimes been considered weevils. Genera of the family have only been found in three formations in
Kazakhstan, with most named in 1993. However, their phylogenetic position is contested, with others considering it part of
Archostemata. The interfamilial relationships of Curculionoidea have been generally well resolved. The
phylogeny by Li et al. (2023) based on phylogenomic data is suggested below: }} ==Families==