Weevils are one of the most diverse groups of animals with thousands of species suspected to remain undiscovered. The Conoderinae Schoenherr, 1833 are no exception, being especially diverse and unknown in the Neotropics where they are recognizable for their unique behaviors and color patterns among weevils. Despite these peculiarities, the group has received little attention from researchers in the past century, with almost nothing known about their evolution. This dissertation presents a series of three studies that begin to elucidate the evolutionary history of these bizarre and fascinating weevils, commencing with an overview of their biology and classificatory history (Chapter 1).

Chapter 2 presents the first formal cladistic analysis on the group to redefine the New World tribes Lechriopini Lacordaire, 1865 and Zygopini, Lacordaire, 1865. An analysis of 75 taxa (65 ingroup) with 75 morphological characters yielded six equally parsimonious trees and synapomorphies that are used to reconstitute the tribes, resulting in the transfer of sixteen genera from the Zygopini to the Lechriopini and four generic transfers out of the Lechriopini to elsewhere in the Conoderinae.

Chapter 3 constitutes a taxonomic revision of the genus Trichodocerus Chevrolat, 1879, the sole genus in the tribe Trichodocerini Champion, 1906, which has had an uncertain phylogenetic placement in the Curculionidae but has most recently been treated in the Conoderinae. In addition to redescriptions of the three previously described species placed in the genus, twenty-four species are newly described and an identification key is provided for all recognized species groups and species.

Chapter 4 quantitatively tests the similarity in color pattern among species hypothesized to belong to several different mimicry complexes. The patterns of 160 species of conoderine weevils were evaluated for 15 categorical and continuous characters. Non-metric multidimensional scaling (NMDS) is used to visualize similarity by the proximity of individual species and clusters of species assigned to a mimicry complex in ordination space with clusters being statistically tested using permutational multivariate analysis of variance (PERMANOVA).