A. Monteiro et Ne. Pierce, Phylogeny of Bicyclus (Lepidoptera : Nymphalidae) inferred from COI, COII,and EF-1 alpha gene sequences, MOL PHYL EV, 18(2), 2001, pp. 264-281
Despite the fact that Bicyclus anynana has become an important model specie
s for wing-pattern developmental biology and studies of phenotypic plastici
ty, little is known of the evolutionary history of the genus Bicyclus and t
he position of B. anynana. Understanding the evolution of development as we
ll as the evolution of plasticity can be attempted in this species-rich gen
us that displays a large range of wing patterns with variable degrees of ph
enotypic responses to the environment. A context to guide extrapolations fr
om population genetic studies within B. anynana to those between closely re
lated species has been long overdue. A phylogeny of 54 of the 80 known Bicy
clus species is presented based on the combined 3000-bp sequences of two mi
tochondrial genes, cytochrome oxidase I and II, and the nuclear gene, elong
ation factor I alpha. A series of tree topologies, constructed either from
the individual genes or from the combined data, using heuristic searches un
der a variety of weighting schemes were compared under the best maximum-lik
elihood models fitted for each gene separately. The most likely tree topolo
gy to have generated the three data sets was found to be a tree resulting f
rom a combined MP analysis with equal weights. Most phylogenetic signal for
the analysis comes from silent substitutions at the third position, and de
spite the faster rate of evolution and higher levels of homoplasy of the mi
tochondrial genes relative to the nuclear gene, the latter does not show su
bstantially stronger support for basal clades. Finally, moving branches fro
m the chosen tree topology to other positions on the tree so as to comply b
etter with a previous morphological study did not significantly affect tree
length. (C) 2001 Academic Press.