Phylogenetic systematics of colotis and associated genera (lepidoptera: pieridae): evolutionary and taxonomic implicationsAccepted on 9 March 2011 2011 Blackwell Verlag GmbH J Zool Syst Evol Res doi: 10.1111/j.1439-0469.2011.00620.x 1Department of Integrative Biology, University of Guelph, Guelph, ON, Canada; 2Jacobys alle 2, 1806 Frederiksberg C, Denmark;3Department of Entomology, Natural History Museum, London, UK; 4Centre de Recherche dÕ Orle´ans, INRA, UR 633 ZoologieForestie re, Orle´ans, France; 5University Museum of Zoology, Cambridge, UK; 657 rue Genot, B-4032 Cheˆne´e, Belgium; 7Kerkhoﬂei54, Hofstade (Zemst), Belgium Phylogenetic systematics of Colotis and associated genera (Lepidoptera: Pieridae):evolutionary and taxonomic implications azrick Nazari *, Torben B. Larsen , David C. Lees , Oskar Brattstro¨m , Thierry Bouyer , Guy Van de Poel and AbstractWe investigated the genetic diversity and phylogenetic placement of the butterﬂies in the genus Colotis and eight related pierid genera usingsequence information from two mitochondrial and two nuclear genes. To establish the status of species, we initially barcoded 632 specimensrepresentative of all genera and most species and subspecies in those genera. A subset was then selected for phylogenetic analysis where additionalgene regions were sequenced: 16S rRNA (523 bp), EF-1a (1126 bp) and wg (404 bp). DNA barcode results were largely congruent with thetraditional classiﬁcation of species in the Colotis group, but deep splits or lack of genetic divergence in some cases supported either species-leveldiﬀerentiation or synonymy. Despite using information from four genes, the deeper nodes in our phylogeny were not strongly supported, andmonophyly of the ÔColotis groupÕ and the genera Colotis and Eronia could not be established. To preserve the monophyly of Colotis, we revive thegenus Teracolus for three outlying species previously in Colotis (i.e. Colotis eris, Colotis subfasciatus and Colotis agoye), as well as the genusAfrodryas for Eronia leda. The position of Calopieris is unresolved although it appears to be well outside the molecular variation in Colotis (s.l.).
A dispersal ⁄ vicariance analysis suggested that major diversiﬁcation in Colotis (s.str.) occurred in Africa with subsequent dispersal to India andMadagascar.
Key words: Pieridae – Colotis – phylogenetic – DNA barcoding – taxonomy in the Pierinae but identiﬁed three genera under the subfamily The Pieridae include some of the most familiar butterﬂies, the ÔTeracolinaeÕ (i.e. Teracolus, Calopieris and Eronia). He also cabbage whites and the grass yellows, yet the apparently long identiﬁed three groups within the genus Teracolus (=Colotis), stable status of many of the species in this family are yet to be one of which he further divided into 12 subgroups based on clariﬁed by molecular means. Several recent papers have used wing colour patterns. Klots (1933) regarded Colotis and Ixias information from gene sequences to shed light on the to be derived from Anthocharidini, considered Eronia, taxonomy or evolutionary history of the family or lower Nepheronia and Pareronia closer to Coliadinae, and expressed ranks therein (Morinaka et al. 2002; Braby 2005; Braby and uncertainties about the placement of Hebomoia and Pinacop- Trueman 2006; Braby et al. 2006, 2007; Chew and Watt 2006; teryx. Talbot (1939), who used the form and colour of the Braby and Pierce 2007; Wheat et al. 2007; Xu et al. 2007; apical spot as the basis for his classiﬁcation, recognized Wheat and Watt 2008; Sua´rez et al. 2009). However, many Gideona as a separate genus and identiﬁed 16 species-groups groups remain largely unexamined. One such group consists of within Colotis, one of which was Colotis (=Calopieris) the genus Colotis and eight closely related genera (i.e. Gideona, eulimene. Peters (1952) also recognized Eronia and Nepheronia Eronia, Nepheronia, Pareronia, Ixias, Pinacopteryx, Hebomoia as separate genera within Teracolinae. Braby et al. (2006) and Calopieris), traditionally placed in diﬀerent tribes within stated, ÔThese (nine) genera may well comprise a separate the subfamily Pierinae but brought together by weakly lineage sister to the rest of Pierinae, but evidence for their supported molecular phylogenies and tentatively named the monophyly is lackingÕ. In case the monophyly of the Colotis ÔColotis groupÕ because of inferred paraphyly (Braby et al.
group is successfully established, the tribal name Teracolini Reuter 1896 is available (Hesselbarth et al. 1995; Braby et al.
There is considerable historical confusion over the status of many taxa in this group, and various revisions present diﬀering The group is entirely Old World and, despite the powerful classiﬁcations and species numbers (Butler 1876; Sharpe 1898– ﬂight ability of some of its members, exhibits considerable 1902; Aurivillius 1925; Talbot 1939; Peters 1952; Ackery et al.
endemism. Hebomoia (2 spp.), Ixias (10 spp.) and Pareronia 1995). Aurivillius (1925) placed ÔHerpaeniaÕ (=Pinacopteryx) (13 spp.) are distributed in India and the Oriental region,Gideona (1 sp.) is endemic to Madagascar, while Calopieris(1 sp.), Eronia (2 spp.), Nepheronia (4 spp.) and Pinacopteryx Corresponding author: Vazrick Nazari (email@example.com) (1 sp.) are conﬁned to Africa and the Arabian Peninsula.
Contributing authors: Torben B. Larsen (torbenlarsen@btinternet.
Calopieris is endemic to the Somali subregion and its only com), David C. Lees (firstname.lastname@example.org), Oskar Brattstro¨m species, Calopieris eulimene, is extremely rare in collections, (email@example.com), Thierry Bouyer (firstname.lastname@example.org), Guy Van de and its position within the Colotis group has not been Poel (email@example.com), Paul D. N. Hebert (firstname.lastname@example.org) substantiated through molecular data (Braby et al. 2006).
With about 46 species, Colotis is a relatively large genus with *Present address: Agriculture and Agri-Food Canada, 3058-C KWNeatby Bldg, 960 Carling Avenue, Ottawa, ON, K1A 0C6 Canada.
a centre of richness in the east African Savannah zone. Outside J Zool Syst Evol Res (2011) 49(3), 204–215 Phylogenetic systematics of Colotis of Africa, ﬁve species are endemic to the Madagascar Sperling 1999; Caterino et al. 2001; Wahlberg et al. 2005; Narita et al.
subregion (i.e. Colotis evanthides, C. evanthe, C. guenei, C.
2006; Braby et al. 2006, 2007; Nazari et al. 2007), and the phylogenetic mananhari and C. zoe), one (C. evagore) occurs as far north as trees were rooted using the Dismorphiinae + Pseudopontiinae clade,the well-established sister to Coliadinae + Pierinae (Braby et al.
southern Spain, many penetrate the Arabian Peninsula and 2006). Because of the narrow focus of the study, no outgroups were Iran, and seven species ﬂy in India and Sri Lanka (i.e. Colotis selected from other lepidopteran families.
amata, C. danae, C. eucharis, C. fausta, C. liagore, C. phisadia Among others, we could not sample the following taxa: Colotis and C. vestalis), where C. etrida and C. protractus are endemic.
eunoma, a rare dune specialist closely related to the C. ione group; Species of Colotis are one of the most prominent insect Nepheronia buquetii buchanani from North Africa and the Arabian components of the savannah zone (sensu Larsen 1984); their Peninsula, C. evagore evagore from Arabia; C. evagore niveus, endemicto the island of Socotra in Yemen and sometimes considered a separate larvae feed on various Capparaceae as well as Salvadora species based on diﬀerences in morphology and genitalia; C. agoye persica (Salvadoraceae), and clouds of them are often seen zephyrus from Somalia; and C. vesta amelia from Western Africa.
around the stands of their food plant or on other ﬂoweringplants. Most Colotis species are morphologically diverse andexhibit pronounced seasonal, sexual and individual variation.
Molecular techniques Wet- and dry-season forms of some species are phenotypically The extraction of total genomic DNA, ampliﬁcation and sequencing very diﬀerent although transitions are common between the were performed in the Biodiversity Institute of Ontario using previ- two extreme seasonal generations (Talbot 1939). Over the ously described protocols (Ivanova et al. 2006). Full-length mtDNA centuries, hundreds of infra-speciﬁc names have been applied barcode sequences (i.e. 658 bp) were obtained for nearly all specimens,and based on results from sequence similarity (Neighbour-Joining) to this wide spectrum of variation, and although many have analyses and the quality of DNA, a subset of specimens was selected been synonymized, some uncertainties remain. Some species for additional gene sequencing (Table S2). Older or failed samples within Colotis have been segregated at subgeneric levels, were targeted using six overlapping primer pairs designed for including C. (Madais) fausta, C. (Cuneacolotis) agoye, C.
cytochrome oxidase I (COI) (Hausmann et al. 2009). Partial sequences (Teracolus) eris and C. (Teracolus) subfasciatus, but these have from 16S ribosomal rRNA and two nuclear genes – Elongation Factor not gained much support. Prior work on wing scale structure, 1-alpha (EF-1a) and wingless (wg) – were also obtained using primersand protocols described previously (Brower and De Salle 1994; Aubert pigmentation and iridescence in Colotis and related genera has et al. 1999). Ampliﬁed DNA from all specimens was bidirectionally also revealed a wide range of invisible (ultraviolet) sexual sequenced for each gene, and ﬁnal sequencing products were run on an dimorphism and patterns invisible to the human eye (Stavenga ABI 3730XL DNA analyzer (Life Technologies, Foster City, CA, and Arikawa 2006; Stavenga et al. 2006; Stavenga and USA). Complementary strands were assembled into contigs and edited Leertouwer 2007; Wijnen et al. 2007), but the lack of phylo- manually, and primers were removed using sequencher 4.5 (Gene genetic hypotheses for the origin and evolution of Colotis and Codes Corporation, Ann Arbor, MI, USA). Sequences were aligned related genera has been a stumbling block in understanding using clustalx 2.0 (Thompson et al. 1997), evaluated by eye andconverted to Nexus using SE-AL 2.0a11 (Rambaut 2002). The 16S how these patterns may have evolved.
rRNA alignment was unambiguous and remarkably variable in the The occurrence of both widespread and narrowly restricted conserved sections. To eliminate noise caused by questionable homol- species of Colotis in three main zoogeographical regions makes ogies in loops in the secondary structure models of the 16S rRNA, we the genus particularly attractive to investigate from a historical retested the data partition after excluding problematic bases without biogeography perspective. The Malagasy and Indian endemics removing doublet characters using the program gblocks 0.91b present several intriguing questions: Was vicariance or dis- (Talavera and Castresana 2007); however, the results were similar.
New sequences were deposited in GenBank, and accession numbers are persal the main process driving evolution of non-African provided in Tables S1 and S2. All data are also available publicly endemic Colotis? Is it possible that, as in some other insects, through the published project ÔCLTÕ on the Barcode of Life Data Colotis originated in Madagascar and spread later to Africa System (bold; http://www.barcodinglife.org).
and India (c.f. Zakharov et al. 2004; Monaghan et al. 2005)?Alternatively, if the ancestors of the non-African Colotisarrived through dispersal from Africa (c.f. Kodandaramaiah Phylogenetic analyses and Wahlberg 2007; Lohman et al. 2008; Aduse-Poku et al.
Neighbour-Joining (NJ) trees for barcode data were constructed 2009), how many dispersal events were involved? In this paper, initially using the quicktree algorithm (Howe et al. 2002) and underthe Kimura two-parameter (K2P) model (Kimura 1980). Additional we (i) use DNA barcoding to test the status of the available taxa NJ and maximum parsimony (MP) analyses were conducted in paup* in Colotis and related genera and (ii) attempt to reconstruct a 4.0 b 10 (Swoﬀord 2002). Heuristic searches for MP analysis were phylogeny for the group with the available molecular data, and carried out with all characters equally weighted and under the tree use it to test the higher level taxonomy and determine the bisection-reconnection (TBR) swapping algorithm with 100 random origins of the genera within the Colotis group.
addition sequences. Bootstrapping of 100 replicates was conductedunder the parsimony criterion with the default setting starting with arandom seed and the TBR branch-swapping algorithm. Bremer Materials and Methods support values were calculated using treerot v.3 (Sorenson andFranzosa 2007). The maximum likelihood (ML) tree was generated using phyml 3.0 online (Guindon and Gascuel 2003), with the A total of 632 specimens representing all but one species in the genus parameters of the best-ﬁt model (GTR + G + I) selected previously Colotis as well as representatives from other genera in the ÔColotis using Multiphyl (Keane et al. 2007) and 100 bootstrap replicates.
groupÕ were examined (Table S1). In most cases, a dry leg accompa- Haplotype diagrams (Figure S1) were constructed in TCS 1.21, with a nied by an image of the specimen was sent to the Biodiversity Institute 90% or 95% conﬁdence limit for parsimony (Templeton et al. 1995).
of Ontario (Guelph, Canada). Vouchers were retained in the originat- Shorter fragments of COI barcodes or those with ambiguous bases ing collections, and the voucher data are publicly available through the were excluded from haplotype analyses.
published project ÔColotis of the WorldÕ (CLT) on the Barcode of Life Bayesian posterior probabilities for the combined data set were ﬁrst Data Systems (BOLD; http://www.barcodinglife.org). Outgroups rep- calculated using mrbayes 3.1.2 (Ronquist and Huelsenbeck 2003) resenting all subfamilies of Pieridae were selected from previous studies under the GTR + G + I model using one cold and three heated and sequences were obtained from GenBank (Table S2; Caterino and simultaneous Markov chain Monte Carlo (MCMC) chains, starting J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH Nazari, Larsen, Lees, Brattstro ¨ m, Bouyer, Van de Poel and Hebert with random initial trees and sampling every 100 generations. The made Pierinae paraphyletic by placing the Coliadinae within analysis was allowed to continue for 5 500 000 generations until the the subfamily. We did not observe a similar pattern in our ML average standard deviation of split frequencies fell below 0.01.
or MrBayes trees, and these were more consistent with Substitution rates were estimated as part of the analysis from default previous studies with better taxon sampling (Wheat et al.
priors, and model parameters were allowed to vary. A total of 5000trees corresponding to the burnin values estimated prior to initiation of each MCMC chain were discarded, and the majority rule consensus The phylogenetic positions of several taxa were unstable tree was generated using the remaining trees with posterior probabil- throughout ML or Bayesian analyses, including C. eulimene, ities plotted on each node.
Hebomoia glaucippe, Ixias pyrene and Eronia leda (Figs 1 and The same data set and model speciﬁcations were used to perform a 2). Across our analyses, Calopieris appeared as either sister to parallel analysis in beast (Drummond and Rambaut 2007; tree not Pierinae, Anthocharidini or Hebomoia with no support. Eronia shown). Branch lengths were allowed to vary according to anuncorrelated lognormal distribution. The Yule process was selected leda strayed from its supposed congener, E. cleodora, and as the tree prior. The MCMC analysis was run twice under default appeared either close to Pareronia or as a sister to the Colotis priors for 10 000 000 generations, sampling the chains every 1000 aurora clade, in both cases with no support. The GenBank generations and yielding a total of 10 000 samples for each run, the COI sequences for Pareronia valeria (AY954573, Braby et al.
ﬁrst 1000 of which were later discarded as burnin. Node posterior 2006) and P. anais (EF584868, Xu et al. 2007) are identical and probabilities and standard deviations were computed for each internal possibly reﬂect misidentiﬁcation.
node using the Tree Annotator module implemented in beast. The None of the gene partitions or the combined analyses estimated posterior probabilities by MrBayes and beast were compa-rable (Table S4). The decay values for each data partition were also recovered the nine genera in the ÔColotis groupÕ as monophy- calculated using treerot and are presented alongside the posterior letic. The genus Colotis was also not monophyletic, with three probabilities in Table S4.
species (i.e. C. agoye, C. subfasciatus and C. eris) consistentlyforming a separate clade, which stayed outside the remainingColotis and closer to Gideona and Pinacopteryx. The mono- Dispersal ⁄ vicariance analysis From the African diversity centre, an African origin for Colotis with [C. eris + (C. subfasciatus + C. agoye)] was supported by later spread to India and Madagascar seems likely. This hypothesis ML analysis, within which several distinct and relatively was tested through a dispersal and vicariance analysis with diva(Ronquist 1997). Given a phylogeny, this method has frequently been well-supported subclades were observed (Fig. 2): used in the reconstruction of ancestral distributions in butterﬂies (e.g.
Group I. etrida, ephyia.
Wahlberg et al. 2005; Kodandaramaiah and Wahlberg 2007; Nazari Group II. aurora, evarne, dissociatus, auxo, incretus.
et al. 2007; Wiemers et al. 2009). Non-Colotis taxa were excluded fromthe Group III. antevippe, rogersi, euippe, pallene, lais, daira, diva analysis because of limited sampling. Unit areas were selected based on biogeographic zones for African butterﬂies (Carcasson 1964; evagore, evanthe, evanthides.
Larsen 1984, 1991). Distributional data for each species were compiled Group IV. liagore, evenina.
in a Nexus ﬁle in mesquite 2.72 (Maddison and Maddison 2009) as Group V. danae, annae (hildebrandti), guenei.
presence ⁄ absence for each region, with the Bayesian phylogeny used Group VI. protractus, fausta, amata, calais, vestalis, for the analysis. Analyses were conducted with and without restriction of maximum number of areas for ancestral nodes. diva assigns a costof zero to vicariance (i.e. allopatric speciation) and duplication (i.e.
Group VII. ungemachi, doubledayi, chrysonome, vesta, sympatric speciation) events and a cost of 1 per unit area to any dispersal and extinction events; thus, the best reconstructions are those Group VIII. zoe, celimene.
that minimize the number of dispersals and extinctions under a Group IX. protomedia, halimede, pleione, venosa, mananhari, parsimony criterion. To avoid accumulation of distribution areas regina, hetaera, elgonensis, ione, erone, (eunoma).
towards the root of the phylogeny, constraints of 2, 3, 4 and 5 unitareas were imposed as the ancestral distribution, and the best The position of C. ungemachi as sister to two clades construction (with the least number of dispersals = 25) was obtained consisting of the more similar C. pleione group (VII) as well when the maximum number of unit areas in ancestral distributions was as the distinct C. zoe + C. celimene group (VIII) was sup- ported only weakly by COI barcodes. Since C. ungemachimorphologically clearly belongs to the pleione (rather than zoe) species group, and considering the lack of nuclear sequences tosupport its current position, we included this species as part of Attempts to sequence the selected genes for all species were (currently paraphyletic) Colotis group VII.
not successful because of a lack of fresh material, and for The MP analysis on EF-1a partition divided the Colotis some species only the COI barcodes were available for group IX into two clades, consisting of (i) C. protomedia, phylogenetic analysis (Table S2). The combined data set C. halimede and C. pleione, and (ii) C. mananhari, C. elgonen- included 2812 positions, of which 1761 were constant, 187 sis, C. ione, C. hetaera and C. regina (Fig. 3). This pattern also were uninformative and 864 were parsimony-informative.
corresponds better with morphology, but it was not supported The wg and the COI barcode partitions had a higher by the other genes or by the combined analysis.
proportion of parsimony-informative characters (39.6% and38.3% respectively) than 16S (25.2%) or EF-1a (26.1%)(Table S3).
Support for the deeper nodes in our phylogenetic trees of the Systematics and biogeography combined data was generally weak or lacking, but many nodeswere consistently recovered through various reconstruction Despite exhaustive taxon sampling of the Colotis group and methods (i.e. MP, ML, Bayesian) (Table S4). Our beast the potential combined support from two nuclear and two phylogeny (not shown) separated Pseudopontiinae from the mitochondrial genes, the lack of resolution at the deeper nodes rest of the taxa with very long branches near the base and of our phylogeny is striking and highlights the possible J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH
Phylogenetic systematics of Colotis L = )33610.72878). Nodes are numbered, and support values are presented in Table S4. Inferred groups of Colotis are identiﬁed with Romannumerals. Red branches show taxa with a varying position between ML and Bayesian phylogenies (Fig. 2) inapplicability of this range of markers for inferring phylog- followed by much longer branches (Kodandaramaiah and enies at least in this group of butterﬂies. Some factors that may Wahlberg 2009; Kodandaramaiah et al. 2010). Some of these explain the low support values include the following: (i) poor eﬀects can be potentially alleviated by adding more basal taxa outgroup sampling; (ii) extremely old age of radiation beyond or removing rogue taxa from the analysis. However, our the range of the markers; (iii) extinction of intermediate outgroup taxa included all major groups of Pieridae, and lineages or species and presence of ﬂoating taxa, like Gideona removal of the rogue taxa (e.g. Hebomoia, Ixias and Calopieris) that reduce the overall support at the base or among the in- did not aﬀect the overall support values in a substantial way group taxa; and (iv) presence of hard polytomies resulting (data not shown). No hard or near-hard polytomies were from rapid speciation, where very short internal branches are evident in our trees, and previous studies employing molecular J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH
Nazari, Larsen, Lees, Brattstro ¨ m, Bouyer, Van de Poel and Hebert Fig. 2. Bayesian tree for Pieridae inferred through MrBayes analysis (Tree Length = 6806 steps, CI = 0.2470, RI = 0.4670, )lnL = )39738.22794). The results of the diva analysis are given for Colotis and related taxa, with maxtrees set to 2 ancestral areas. Coloured cladesindicate Colotis groups identiﬁed through this study, and red branches highlight taxa with a varying position between Bayesian and maximumlikelihood phylogenies (Fig. 1). Figured species are, from top to bottom, Calopieris eulimene, Hebomoia glaucippe, Nepheronia argia, Pinacopteryxeriphia, Teracolus eris, Colotis incretus, Colotis evanthe evanthides, Colotis evenina, Colotis amata, Colotis ungemachi, Colotis zoe and Colotis ione clock methods do not report an extremely old age for the The position of the Malagasy genus Gideona as sister to group (Braby et al. 2007; Wheat et al. 2007).
Pinacopteryx + Teracolus, although weakly supported, is The position of Calopieris as sister to Eronia (Fig. 1), consistent with previous ﬁndings (Braby et al. 2006). The Pierinae (Fig. 2) or Hebomoia + Anthocharidini (beast, not adult of Gideona lucasi is superﬁcially reminiscent of Hebo- shown) is also not well supported, but it strongly suggests that moia, and in the past this species has been included in the the morphologically distinct Calopieris belongs well outside genus Colotis. Our diva analysis suggests that the last common Colotis sensu lato. Similarly, Ixias and Hebomoia have ancestor of this group with Ixias and Eronia ﬂew in Africa, and unstable positions across our molecular reconstructions. The subsequently dispersed to Madagascar (Gideona) and India deep divergence between E. leda and E. cleodora barcodes (Ixias, Eronia). Independent dispersals to Madagascar are also (15.9 ± 0.21%), their completely diﬀerent colour patterns and noted for Nepheronia pauliani and C. mananhari, possibly their paraphyly strongly suggest that these two species do not much later than Gideona. Most other Malagasy Colotis show belong in the same genus.
shallow divergence from their sister taxa and hence seem to J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH Phylogenetic systematics of Colotis Fig. 3. Consensus trees resulting from Maximum Parsimony analysis for each gene partition. SC = strict consensus of (n) trees, TL = treelength, CI = consistency index, RI = retention index have arrived on the island more recently, including C. calais within some species supported their separate species status, crowleyi and C. evanthides, derived from common ancestors in including the Malagasy N. buquetii pauliani (9.5 ± 0.3%) and Africa (C. calais) or Madagascar (C. evanthe). Multiple inde- Pinacopteryx eriphia mabillei (5.8 ± 0.3%), as well as the pendent dispersals from Africa to Madagascar have been Tanzanian Eronia cleodora (4.6 ± 0.9%) restricted to the documented in other butterﬂies, including those in the genus coastal forests in eastern Kenya and Tanzania and with a Charaxes (Aduse-Poku et al. 2009).
much wider black border on its wings.
In all of our phylogenies, the taxa eris, subfasciatus and The divergence between the two subspecies of Teracolus agoye always formed a well-supported clade outside the agoye (i.e. ssp. agoye and spp. bowkeri, 2.0 ± 0.3%) supports Colotis, usually close to Pinacopteryx and Gideona. Shared the traditional status of these taxa (Larsen 1992). No subspe- diﬀerences between these species and other Colotis have been cies are currently recognized under Teracolus eris; however, the noted, including an acute forewing apex, a shorter aedeagus presence of several distinct mitochondrial lineages within eris and a produced valval apex (Henning et al. 1997), as well as suggests an overlooked geographic structure that requires diﬀerences in larval morphology (Larsen 1992). A separate further investigation (Fig. 4).
generic status for these three species is required.
DNA barcode divergence between C. auxo (South Africa) Both Indian endemic species, C. aurora (group I) and and C. dissociatus (Botswana) is very shallow (1.0 ± 0.1%).
C. etrida (group II), seem to have common ancestors with These two live in very diﬀerent habitats in South Africa with African species (C. evarne and C. ephyia, respectively) which no overlap: while auxo occupies the coastal forests and the range as far north as Sudan, Chad, Saudi Arabia and Yemen bushveld, dissociatus mainly thrives in savannah habitats (TBL (Fig. 2). The shallow divergence between these species and observation). Both have a white to pale yellow ground colour, their presence in the Arabian Peninsula rejects the possibility with dissociatus being smaller, generally paler and often with of an old vicariance speciation resulting from the severance of no trace of black margin on the inner edge of the orange tip of the Indian plate from Africa or Madagascar and instead the forewings. Our C. auxo from Kenya ⁄ Tanzania (i.e. ssp.
suggests a more recent vicariance or dispersal event in the Near incretus) were signiﬁcantly divergent from both of these East that resulted in isolation of C. etrida and C. aurora in the (11.2 ± 0.1%). These are usually larger with much deeper yellow ground colour and more pointed forewings.
With Teracolus removed, and ignoring the unsupported Colotis phisadia and C. vestalis are two closely related position of Afrodryas leda in our Bayesian phylogeny, Colotis species often distinguishable by their forewing ground colour forms a monophyletic clade that is primarily supported by the (i.e. pink-salmon in C. phisadia, white in C. vestalis). They EF-1a gene (Figs 1 and 2; Table S4). Despite unique distribu- share the same larval host (Salvadora persica, Nazari 2003) and tions and several autapomorphies, the status of Madais occur sympatrically from Iran to Pakistan and Gujarat in Swinhoe (1909) or Cuneacolotis Henning et al. (1997) is not India. Colotis phisadia is a migratory species (Gardner and supported as subgenera under Colotis. Instead, we prefer to Howarth 2007) and extends into Arabia and Africa, while recognize Ôspecies groupsÕ that formed well-supported clusters.
C. vestalis spreads deeper into Pakistan and wet habitats as faras Delhi; it is conspicuously absent from Arabia but re-appearsin the dry parts of East Africa with a radically diﬀerent DNA barcoding and taxonomy morphology (i.e. ssp. castalis). With the exception of the latter DNA barcoding (Hebert et al. 2003, 2004; Hajibabaei et al.
population, all other phisadia and vestalis examined in our 2006) is rapidly becoming the most important tool for species study were nearly identical in their DNA barcodes and shared identiﬁcation and discovery. Application of DNA barcoding haplotypes (Figure S1). Intermediate phenotypes, often with a to assess mitochondrial sequence diversity among material yellow ground colour, are not uncommon when the two occur examined in our study highlighted the inconsistencies between in sympatry (e.g. in Southern Iran; see Nazari 2003). Colotis current taxonomy and genetic variation in many of the species.
protractus, traditionally considered a subspecies of C. phisadia, Divergent populations showing deep splits (>4%) observed occurs sympatrically with vestalis as well as phisadia in J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH Nazari, Larsen, Lees, Brattstro ¨ m, Bouyer, Van de Poel and Hebert Fig. 4. Neighbour-joining trees of the cytochrome oxidase I barcodes for Colotis (a) and associated genera (b). Taxa with changed ranks orcombinations are in bold Southern Iran to Gujarat and is immediately distinguishable Angola and Botswana, and (iii) Somalia and Ethiopia. A by the pink-salmon ground colour on both wings and blue similar situation in C. euippe is taxonomically more diﬃcult to forewing apical spots. Its barcode also clearly separates it as a explain: while individuals from Gambia, Nigeria, Cameroon distinct species closer to C. fausta (Fig. 4).
and Somalia form a cluster (II), those from Congo, Angola, Divergent clusters within C. evagore demonstrate a geo- Namibia, Tanzania and Uganda form two separate clusters graphic structure, with populations in (i) Spain, Morocco, (I, III) with geographic overlap. Minor but consistent diﬀer- Tunisia and Nigeria, corresponding to ssp. nouna, (ii) Congo, ences in coloration of the forewing tip as well as the underside J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH Phylogenetic systematics of Colotis of the wings in these two clusters suggests the possibility of an the Kenya (Coast) (3.3 ± 0.2%) is comparable to the distance overlooked species, but further investigation is needed (TB from their closest sister species, C. erone (i.e. 3.7% and 3.6%, observation). Our specimens of C. rogersi did not show any respectively), which supports a distinction at species level. No barcode diﬀerences from C. euippe; however, misidentiﬁca- subspecies are currently recognized under C. ione (Ackery tions could not be ruled out.
et al. 1995). Further sampling of populations (e.g. from South The Congolese C. pallene, with uniquely wide submarginal Africa) is needed before a third species is recognized in this bands on the upperside of the hindwings and a notable The type locality of C. hetaera is ÔEndaraÕ (i.e. Mt. Ndara in (3.2 ± 0.8%), presents another case of a potentially over- Coast, Kenya, not Zanzibar as indicated by Ackery et al.
looked species ﬂagged by DNA barcoding. Examination of 1995). Several individuals of C. hetaera, including a yellow additional material from South Africa and Tanzania is female (GVDP087) from the Golini Forest in Kenyan Coast required before a separate status for this population can be (I), were divergent from other subspecies (i.e. aspasia, ankol- ensis and lorti) (II). However, a white female from the same Despite a wide range geographic sampling of C. antevippe, forest (GVDP082) fell within the second group. The males in small barcode variation observed among these populations the two clusters are similar. This indicates the possible (1.06 ± 0.04%) does not support the recognition of subspecies coexistence of a cryptic species sympatric with C. hetaera in within C. antevippe. The stability of mitochondrial DNA over the Kenyan Coast.
a large area with a high climatic variation can be explained bya rapid recent range expansion, a phenomenon that deservesfurther investigation.
Colotis evanthides is a rare species conﬁned to the Comoros, Our study highlights the paraphyly of the nine genera currently Aldabra, Assumption, Cosmoledo and Astove Islands in the known as the Colotis group and underscores the need for Indian Ocean. On the basis of similarities in genitalia, additional gene sampling to resolve the phylogenetic relation- Bernardi (1954) hypothesized a close aﬃnity between C. e- ships within the family Pieridae. Major diversiﬁcation in vanthides and the Malagasy C. evanthe, the African C. aurora Nepheronia, Pinacopteryx and Colotis seems to have occurred and the Indian C. etrida. Thus, C. evanthides was later termed in Africa during the Eocene, while Hebomoia, Ixias and Ôthe Lemurian linkÕ (Cogan and Hutson 1971). However, our Pareronia diversiﬁed in the Oriental region. The Malagasy phylogenetic analysis does not support these four species as a species seem to represent a combination of old and new natural group.
arrivals to the Island, while the Indian endemics appear to be Within C. celimene, individuals from Nigeria (ssp. sudani- derived from common ancestors that lived in the area between cus), Niger, Congo and Kenya show only minor variation in northeast Africa and Arabia to India.
their barcodes (0.3 ± 0.1%). Several subspecies currently DNA barcoding largely supported the traditional species recognized under C. celimene (e.g. sudanicus, amina, angusi, taxa in Colotis and related genera, but we found evidence for etc.) may be redundant; however, the Namibian ssp. pholoe is exclusion of three species (i.e. C. eris, C. subfasciatus and divergent (2.5 ± 0.2%). The Somali ssp. praeclarus has been C. agoye) from the genus Colotis and the taxon leda from recently proposed as a distinct species (Bouyer 2010).
Eronia, for which we assign new generic status (i.e. Teracolus Colotis chrysonome, C. aurigineus, and C. vesta are closely and Afrodryas). Several cases of undetected species-level related species with overlapping ranges in eastern Africa.
variation, or lack thereof, were also ﬂagged by DNA barcod- Colotis doubledayi, a rare species conﬁned to the Namib dry ing, where relevant taxonomic changes are proposed. To detect zone in southern Africa, is another member of this group with such taxonomic discrepancies, future studies should also aim a distinct barcode haplotype (Fig. 4). Within the widespread for even more comprehensive population sampling, particu- C. chrysonome, we observed gaps (3.3 ± 0.5%) between larly in the genera Ixias and Pareronia that were sparsely populations from Sudan (type locality), Kenya and Ethiopia, sampled in our study.
suggesting undetected diﬀerentiation (Fig. 4). The often largerand variable C. vesta formed two separate clusters withspecimens from Congo (Katanga) appearing in both. One of these clusters (i.e. Kenya, Tanzania and Congo) also included We thank Jean-Francois Landry, Don Lafontaine, and Jeremy the sympatric C. aurigineus (i.e. Kenya, Tanzania and Ugan- DeWaard (Canada), Robert Robbins and Claire Kremen (USA),John Tennent and Keith Stiﬀ (UK), Rodolphe Rougerie and Isabelle da) (Fig. 4). Presence of shared haplotypes between these Meusnier (France), Roger Vila (Spain), Christopher Wheat (Finland), otherwise readily recognizable sister species when in sympatry Robin van Velzen and Hein Leertouwer (the Netherlands), Alison suggests gene ﬂow between the two (Figure S1).
Cameron (Ireland), Wolfgang ten Hagen (Germany), Alireza Naderi Within the montane species C. elgonensis, the disjunct ssp.
(Iran), Steve Collins (Kenya), and Gerardo Lamas (Peru) for providing glauningi from Nigeria seems to be distinct. The Congo specimens or support. Doekele G. Stavenga (Groningen, the Nether- (Katanga) ssp. nobilis also shows some diﬀerentiation; how- lands) generously provided his data and commentary on UV reﬂection ever, it falls within a cluster with all other populations from patterns in Pierinae, which will be published separately. We also thankthe staﬀ at the NHM (UK), RMCA (Belgium) and Naturalis (the Uganda, Burundi, Kenya and Congo (Kivu Nord), including Netherlands) Museums and the Royal Antwerp Entomological Society ssp. elgonensis and ssp. basilewskyi from Uganda and ssp.
(Belgium). We also thank Michael Braby (Australia), Niklas Wahlberg (Finland) and anonymous JZSER reviewers who provided helpful (0.5 ± 0.3%). The three montane distribution areas are well comments on the earlier drafts of this manuscript. DCL was supported separated from each other, with the Nigerian population by a STUDIUM fellowship and INRA. Some Malagasy collections isolated by over 2000 km from Kivu.
were collected under NSF grant DEB-0072713 to B. L. Fisher andC. E. Griswold. This research was supported through funding to the The distance between C. ione (I) from Somalia, Tanzania Canadian Barcode of Life Network from Genome Canada (through and Kenya (Eastern, Nyaza) and those (II) from Congo and J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH Nazari, Larsen, Lees, Brattstro ¨ m, Bouyer, Van de Poel and Hebert the Ontario Genomics Institute), NSERC and other sponsors listed at Braby MF, Pierce NE, Vila R (2007) Phylogeny and historical biogeography of the subtribe Aporiina (Lepidoptera: Pieridae):implications for the origin of Australian butterﬂies. Biol J Linn Soc90:413–440.
Brower AVZ, De Salle R (1994) Practical and theoretical consider- ations for choice of a DNA sequence region in insect molecular Syste´matique et Phylogenie de Colotis et des genres apparente´s systematics, with a short review of published studies using nuclear (Le´pidopte res: Pieridae): implications e´volutives et taxonomiques gene regions. Ann Entomol Soc Am 87:702–716.
Nous avons e´tudie´ la diversite´ ge´ne´tique et la position phyloge´ne´tique Butler AG (1869) Descriptions of new Rhopalocera from the collection des papillons du genre Colotis et de huit autres genres de Pie´ride´s of Herbert Druce Esq. Cistula Entomol 1:1–16.
apparente´s, a lÕaide de quatre ge nes, deux mitochondriaux et deux Butler AG (1876) Revision of the lepidopterous genus Teracolus nucle´aires. Aﬁn dÕe´tablir le statut taxonomique des espe ces, nous avons with descriptions of the new species. Proc Zool Soc Lond 1876:126– dans un premier temps e´tabli les codes-barres ge´ne´tiques de 523 spe´cimens repre´sentant tous les genres et la plupart des espe ces et sous- Carcasson RH (1964) A preliminary survey of the zoogeography of espe ces constituantes. Nous avons ensuite choisi un sous-ensemble de African butterﬂies. East Afr Wildl J 2:122–157. [reprinted in Ackery ces taxons que nous avons soumis a une analyse phyloge´ne´tique apre s et al. 1995].
avoir e´tabli les se´quences de portions des ge nes suivants : 16S ARNr Caterino MS, Sperling FAH (1999) Papilio phylogeny based on (523 bp), EF-1a (1126 bp) et wg (404 bp). Les re´sultats des codes- mitochondrial cytochrome oxidase I and II genes. Mol Phylogenet barres ge´ne´tiques concordent bien dans lÕensemble avec la classiﬁcation Evol 11:122–137.
traditionnelle des espe ces du groupe Colotis, mais quelques cas de Caterino MS, Reed RD, Kuo MM, Sperling FAH (2001) A divergence ge´ne´tique prononce´e, ou a lÕinverse des cas de similitudes partitioned likelihood analysis of swallowtail butterﬂy phylogeny ge´ne´tiques, indiquent soit des diﬀe´renciations spe´ciﬁques insoup- (Lepidoptera: Papilionidae). Syst Biol 50:106–127.
c¸onne´es, soit des cas de synonymie. Malgre´ lÕanalyse de donne´es Chew FS, Watt WB (2006) The Green-veined White (Pieris napi L.), provenant de quatre ge nes, les branchements de base de notre arbre its pierine relatives, and the systematics dilemmas of divergent phyloge´ne´tique sont peu ﬁables et nous nÕavons pas re´ussi a e´tablir la character sets (Lepidoptera, Pieridae). Biol J Linn Soc 88:413– monophylie du Ôgroupe ColotisÕ, ni des genres Colotis et Eronia. Aﬁn de maintenir la monophylie de Colotis, nous avons re´tabli le genre Cogan BH, Hutson AM (1971) Preliminary observations on the Teracolus pour y transfe´rer trois espe ces de Colotis (Colotis eris, afﬁnities and composition of the insect fauna of Aldabra. Phil Trans Colotis subasciatus et Colotis agoye), et le genre Afrodryas pour inclure R Soc Lond B Biol Sci 260:315–325.
E. leda. La position de Calopieris nÕest pas re´solue mais ce genre Cramer P (1780) De Uitlandsche Kapellen voorkomende in de drie pre´sente un proﬁl bien au-dehors de la variation observe´e dans Colotis waereld-deelen Asia, Africa en America. Amsteldam and Utrecht.
(s.l.). Une analyse de vicariance ⁄ dispersion indique quÕune diversiﬁca- 4 [part]:1–28.
tion importante des Colotis (s. str.) aurait eu lieu en Afrique, suivie Drummond AJ, Rambaut A (2007) BEAST: Bayesian evolutionary dÕune dispersion vers lÕInde et Madagascar.
analysis by sampling trees. BMC Evol Biol 7:214.
Drury D (1773) Illustrations of Natural History. White, London.
Fabricius JC (1775) Systema Entomologiae. Flensburgi and Lipsiae.
Gardner D, Howarth B (2007) A butterﬂy migration: blue-spotted Arab at Jabal Barakah, western Abu Dhabi. Tribulus 17:98.
Ackery PR, Smith CR, Vane-Wright RI (1995) CarcassonÕs African Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm to Butterﬂies: An Annotated Catalogue of the Papilionoidea and estimate large phylogenies by maximum likelihood. Syst Biol Hesperioidea of the Afrotropical Region. British Museum (Natural History), London, UK.
Hajibabaei M, Janzen DH, Burns JM, Hallwachs W, Hebert PDN Aduse-Poku K, Vingerhoedt E, Wahlberg N (2009) Out-of-Africa (2006) DNA barcodes distinguish species of tropical Lepidoptera.
again: a phylogenetic hypothesis of the genus Charaxes (Lepidop- Proc Natl Acad Sci USA 103:968–971.
tera: Nymphalidae) based on ﬁve gene regions. Mol Phylogenet Evol Hausmann A, Hebert PDN, Mitchell A, Rougerie R, Sommerer M, Edwards T, Young CJ (2009) Revision of the Australian Oenochro- Aubert J, Legal L, Descimon H, Michel F (1999) Molecular phylogeny ma vinaria Guene´e, 1858 species-complex (Lepidoptera: Geometri- of swallowtail butterﬂies of the tribe Papilionini (Papilionidae, dae, Oenochrominae): DNA barcoding reveals cryptic diversity and Lepidoptera). Mol Phylogenet Evol 12:156–167.
assesses status of type specimen without dissection. Zootaxa 2239:1– Aurivillius C (1925) Pieridae. In: Seitz A (ed.), Die Gross-Schmetter- linge der Erde. 13. Band: Die Afrikanischen Tagfalter. Alfred Hebert PDN, Cywinska A, Ball SL, Dewaard JR (2003) Biological Kernen, Stuttgart, pp 29–69.
identiﬁcations through DNA barcodes. Proc R Soc Lond B Biol Sci Bernardi G (1954) Notes sur la taxonomie de deux Colotis malgaches.
Le Naturaliste malgache 6:57–60.
Hebert PDN, Penton EH, Burns JM, Janzen DH, Hallwachs W (2004) Boisduval JBA (1847) Catalogue des Le´pidopte res recuellis par M.
Ten species in one: DNA barcoding reveals cryptic species in the Delegorgue pendant les anne´es 1838–1844 a Port-Natal, au pays de neotropical skipper butterﬂy Astraptes fulgerator. Proc Natl Acad Amazoulous et dans la contre´e de Massilicatzi. In: Delegorgue A Sci USA 101:14812–14817.
(ed.), Voyage dans lÕAfrique australe, 2:585–602.
Hecq J (1975) Le complexe Colotis eucharis F. (Lepidopt. Pieridae).
Bouyer T (2010) Colotis praeclarus (Butler, 1886) (Lepidoptera, Pieridae). Entomologia Africana 15:22–24.
Henning GA, Henning SF, Joannou JG, Woodhall SE (1997) Living Braby MF (2005) Afrotropical mistletoe butterﬂies: larval food plant Butterﬂies of the Southern Africa. Umdaus Press, Hatﬁeld, South relationships of Mylothris Hu¨bner (Lepidoptera: Pieridae). J Nat Hist 39:499–513.
Hesselbarth G, Van Oorschot H, Wagener S (1995) Die Tagfalter der Braby MF, Pierce NE (2007) Systematics, biogeography and diversi- Tu¨rkei, I. Selbstverlag Sigbert Wagener, Bocholt.
ﬁcation of the Indo-Australian genus Delias Hu¨bner (Lepidoptera: Howe K, Bateman A, Durbin R (2002) QuickTree: building huge Pieridae): phylogenetic evidence supports an Ôout-of-AustraliaÕ origin. Syst Entomol 32:2–25.
Braby MF, Trueman JWH (2006) Evolution of larval host plant Hu¨bner J (1823) Sammlung exotischer Schmetterlinge, 2. In: Hu¨bner J associations and adaptive radiation in pierid butterﬂies. J Evol Biol (1819–1827). Jacob Hu¨bner Verlag, Augsburg.
Ivanova NV, Dewaard JR, Hebert PDN (2006) An inexpensive, Braby MF, Vila R, Pierce NE (2006) Molecular phylogeny and automation-friendly protocol for recovering high-quality DNA. Mol systematics of the Pieridae (Lepidoptera: Papilionoidea): higher Ecol Notes 6:998–1002.
classiﬁcation and biogeography. Zool J Linn Soc 147:239–275.
J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH Phylogenetic systematics of Colotis Keane TM, Naughton TJ, McInerney JO (2007) MultiPhyl: a high- Staudinger O (1884) Exotische Tagfalter in systematischer Reinfolge throughput phylogenomics webserver using distributed computing.
mit Beru¨cksichtigung neuer Arten. In: Staudinger O, Schatz E (eds), Nucleic Acids Res 35:W33–W37.
Exotische Schmetterlinge. 4, Theil I, VII, pp 43–44.
Kimura M (1980) A simple method for estimating evolutionary rates Stavenga DG, Arikawa K (2006) Evolution of color and vision of of base substitutions through comparative studies of nucleotide butterﬂies. Arthropod Struct Dev 35:307–318.
sequences. J Mol Evol 16:111–120.
Stavenga DG, Leertouwer HL (2007) Colourful butterﬂy wings: scale Klots AB (1933) A generic revision of the Pieridae (Lepidoptera) stacks, iridescence and sexual dichromatism of Pieridae. Entomol together with a study of the male genitalia. Entomol Am (NS) Ber (Amsterdam) 67:158–164.
Stavenga DG, Giraldo MA, Hoenders BJ (2006) Reﬂectance and Klug JCF (1829) . Symbolae physicae seu icones et descriptiones transmittance of light scattering scales stacked on the wings of pierid insectorum quae ex itinere per Africam borealem et Asiam occiden- butterﬂies. Optics Express 14:4880–4890.
talem Fridrichi Guilelmi Hemprich et Christiani Godofredi Ehren- Stoneham HF (1957) Two new generic names, and new subspecies and berg Medicinae et Chirurgiae Doctorum studio novae aut illustratae forms of Pieridae. Bull. Stoneham Mus. 70:1–3.
redierunt. Decas Prima (1929), Berolini: ex Oﬃcina Academia, Sua´rez NM, Betancor E, Pestano J (2009) Intraspeciﬁc evolution of Impensis G. Reimer.
Canarian Euchloe (Lepidoptera: Pieridae) butterﬂies, based on Kodandaramaiah U, Wahlberg N (2007) Out-of-Africa origin and mtDNA sequences. Mol Phylogenet Evol 51:601–605.
dispersal mediated diversiﬁcation of the butterﬂy genus Junonia Swainson W (1833) (ed.) Zoological illustrations, or original ﬁgures (Nymphalidae: Nymphalinae). J Evol Biol 20:2181–2191.
and descriptions of new, rare or interesting animals selected chieﬂy Kodandaramaiah U, Wahlberg N (2009) Phylogeny and biogeography from the classes of ornithology, entomology and conchology, (2) 3, of Coenonympha butterﬂies (Nymphalidae: Satyrinae) – patterns of [ix]. James Moyes, London.
colonization in the Holarctic. Syst Entomol 34:315–323.
Swinhoe C (1909) New species of Indo-Malayan and African Kodandaramaiah U, Pen˜a C, Braby MF, Grund R, Mu¨ller CJ, Nylin Lepidoptera. Ann Mag Nat Hist 3:89–98.
S, Wahlberg N (2010) Phylogenetics of Coenonymphina (Nymphal- Swoﬀord DL (2002) PAUP*. Phylogenetic Analysis Using Parsimony idae: Satyrinae) and the problem of rooting rapid radiations. Mol (*and Other Methods). Version 4. Sinauer Associates, Sunderland, Phylogenet Evol 54:386–394.
Larsen TB (1984) Butterﬂies of Saudi Arabia and its Neighbours.
Talavera G, Castresana J (2007) Improvement of phylogenies after Stacey International, London.
removing divergent and ambiguously aligned blocks from protein Larsen TB (1991) The Butterﬂies of Kenya and their Natural History.
sequence alignments. Syst Biol 56:564–577.
Oxford University Press, Oxford.
Talbot G (1939) Revisional notes on the genus Colotis Hubn. (Lepid.
Larsen TB (1992) On the status of Colotis agoye (Wallengren, 1857) Pieridae) with a systematic list. Trans R Entomol Soc Lond 88:173– and its subspecies. Metamorphosis 3:5–10.
Larsen TB (2005) Butterﬂies of West Africa. Apollo Books, Stenstrup, Templeton AR, Routman E, Phillips CA (1995) Separating population structure from population history: a cladistic analysis of the Lohman DJ, Peggie D, Pierce NE, Meier R (2008) Phylogeography geographical distribution of mitochondrial DNA haplotypes in the and genetic diversity of a widespread Old World butterﬂy, Lampides tiger salamander, Ambystoma tigrinum. Genetics 140:767–782.
boeticus (Lepidoptera: Lycaenidae). BMC Evol Biol 8:301.
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG Maddison WP, Maddison DR (2009) Mesquite: a modular system for (1997) The CLUSTALX Windows inference: ﬂexible strategies for evolutionary analysis. Version 2.72. Available from: http://mes- multiple sequence alignment aided by quality analysis tools. Nucleic quiteproject.org (accessed December 2009).
Acid Res 25:4876–4882.
Monaghan MT, Gattolliat JL, Sartori M, Elouard JM, James H, Wahlberg N, Braby MF, Brower AVZ, de Jong R, Lee MM, Nylin S, Derleth P, Glaizot O, de Moor F, Vogler AP (2005) Trans-oceanic Pierce N, Sperling FAH, Vila R, Warren AD, Zakharov E (2005) and endemic origins of the small minnow mayﬂies (Ephemeroptera, Synergistic effects of combining morphological and molecular data Baetidae) of Madagascar. Proc R Soc Lond B Biol Sci 272:1829– in resolving the phylogeny of butterﬂies and skippers. Proc R Soc Lond B Biol Sci 272:1577–1586.
Morinaka S, Miyata T, Tanaka K (2002) Molecular phylogeny of the Wallengren HDJ (1857) Kafferlandets Dag-fja¨rilar, insamlade a˚ren Eichhorni group of Delias Hu¨bner, 1819 (Lepidoptera: Pieridae).
1838–1845 af J. A. Wahlberg ⁄ Lepidoptera Rhopalocera in Terra Mol Phylogenet Evol 23:276–287.
Caffrorum annis 1838–1845 collecta a J. A. Wahlberg. K Svenska Narita S, Nomura M, Kato Y, Fukatsu T (2006) Genetic structure of Vetensk Akad Handl 2:1–55.
sibling butterﬂy species affected by Wolbachia infection sweep: Wheat CW, Watt WB (2008) A mitochondrial-DNA-based phylo- evolutionary and biogeographical implications. Mol Ecol 15:1095– geny for some evolutionary-genetic model species of Colias butterﬂies (Lepidoptera, Pieridae). Mol Phylogenet Evol 47:893– Nazari V (2003) Butterﬂies of Iran. Dayereye Sabz publications, Wheat CW, Vogel H, Wittstock U, Braby MF, Underwood D, Nazari V, Zakharov EV, Sperling FAH (2007) Phylogeny, historical Mitchell-Olds T (2007) The genetic basis of a plant–insect coevo- biogeography, and taxonomic ranking of Parnassiinae (Lepidoptera, lutionary key innovation. Proc Natl Acad Sci USA 104:20427– Papilionidae) based on morphology. Mol Phylogenet Evol 42:131– Wiemers M, Keller A, Wolf M (2009) ITS2 secondary structure Peters W (1952) A Provisional Checklist of the Butterﬂies of the improves phylogeny estimation in a radiation of blue butterﬂies of Ethiopian Region. E.W. Classey, Feltham, Middlesex.
the subgenus Agrodiaetus (Lepidoptera: Lycaenidae: Polyommatus).
Rambaut A (2002) Sequence alignment editor, version 2.0. Avail- BMC Evol Biol 9:300.
Wijnen B, Leertouwer HL, Stavenga DG (2007) Colors and pterin (accessed December 2009).
pigmentation of pierid butterﬂy wings. J Insect Physiol 53:1206– Ronquist F (1997) DIVA (dispersal-vicariance analysis) ver. 1.2.
Computer program and manual. Available from http://www.ebc.
Xu L, Hao JS, Zhu GP, Yin XB, Pan HC, Huang DY, Zhang XP uu.se/systzoo/research/diva/diva.html (accessed January 2010).
(2007) Molecular phylogenetic relationships of some species and Ronquist F, Huelsenbeck JP (2003) MRBAYES 3: Bayesian phyloge- genera in Pierinae and Coliadnae [Coliadinae] (Pieridae) based on netic inference under mixed models. Bioinformatics 19:1572–1574.
partial sequence of mitochondrial CO I and Cyt b genes. Acta Sharpe EM (1898–1902) Monographiae Entomologicae. 1. A Mono- Zootaxon Sinica 32:842–850.
graph of the Genus Teracolus. Lovell Reeve, London.
Zakharov EV, Smith CR, Lees DC, Cameron A, Vane-Wright RI, Sorenson MD, Franzosa EA (2007) TREEROT, Version 3. Boston Sperling FAH (2004) Independent gene phylogenies and morphol- University, Boston, MA. Available from http://people.bu.edu/mso- ogy demonstrate a Malagasy origin for a wide-ranging group of ren/TreeRot.html (accessed December 2009).
swallowtail butterﬂies. Evolution 58:2763–2782.
J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH Nazari, Larsen, Lees, Brattstro ¨ m, Bouyer, Van de Poel and Hebert (9) Colotis evanthe and C. evanthides are so similar in Appendix: Proposed taxonomic changes morphology as well as DNA barcodes (1.11 ± 0.04%) that (1) The taxon leda was originally described in the genus Dryas we suggest the name evanthides be used as a subspecies of (Boisduval 1847), a homonym of Dryas Hu¨bner 1823; C. evanthe (stat. nov.).
currently the valid genus for Dryas iulia (Heliconiinae). Butler (10) Idmais vesta Reiche, (1850) is a junior primary homonym (1869) subsequently placed leda under Eronia Hu¨bner, 1823.
of Idmais vesta Boisduval (1847). Under Art. 23.9.2 of the Afrodryas Stoneham (1957) was described as a replacement International Code for Zoological Nomenclature, to maintain name for Dryas Boisduval (1847); but later treated as a usage of vesta Reiche, the name Idmais vesta Boisduval (1847) synonym of Eronia. Our results support the arrangement for needs to be designated as a Ônomen oblitumÕ, and Idmais vesta leda by Boisduval (1847) and Stoneham (1957). Accordingly, Reiche, (1850) as a Ônomen protectumÕ. This taxonomic act we revive the genus Afrodryas and the combination Afrodryas shall be properly completed in a separate publication.
leda (comb. rev.).
(2) We found that Colotis agoye, C. subfasciatus and C. erisconsistently formed a well-supported clade separate and Revised list of species and subspecies outside all other Colotis. Here, we revive the oldest available Unexamined taxa are marked by an asterisk (*). The Colotis generic name for these species, Teracolus Swainson, (1833) groups I–IX are listed according to the results of this study.
(stat. rev.) (Type species: Papilio subfasciatus) to represent this For additional synonymy and references to original descrip- tions, see Ackery et al. (1995).
(3) We also recognize the divergent populations Nepheronia (1) Genus Hebomoia Hu¨bner, (1819) pauliani (stat. nov.), Pinacopteryx mabillei (stat. nov.) and H. glaucippe (Linnaeus, 1758) Eronia dilatata (stat. rev.) as good species.
H. leucippe (Cramer, 1775)* (4) The taxon incretus was originally described as a separate (2) Genus Gideona Klots, 1933 species based on a single female from Mamboia (Tanzania) G. lucasi (Grandidier, 1867) and later treated as a subspecies of Colotis auxo (Hecq 1975).
(3) Genus Ixias Hu¨bner, (1819) Examination of the holotype in the NHM conﬁrmed the I. ﬂavipennis Grose-Smith, 1885* identity of our divergent Kenya ⁄ Tanzania specimens. We I. kuehni Ro¨ber, 1891* therefore reinstate C. incretus as a valid species and consider I. malumsinicum Thieme, 1897* dissociatus an ecological (savannah) subspecies of C. auxo I. marianne (Cramer, 1779) (stat. nov.).
I. paluensis Martin, 1914* (5) The correct name for the Indian species (Papilio) eucharis I. piepersii (Snellen, 1877)* (Fabricius 1775) (junior primary homonym of Delias eucharis I. pyrene (Linnaeus, 1764) Drury 1773) is Colotis aurora (Cramer 1780; Larsen 2005).
I. reinwardtii (Vollenhoven, 1860)* Considering the large gap between the Indian and the African I. venilia (Godart, 1819)* C. aurora (5.34 ± 0.03%), we recognize the African popula- I. vollenhovii (Wallace, 1867)* tion as a separate species, C. evarne (Klug 1829) (stat. rev.).
(4) Genus Calopieris Aurivillius, 1899 (6) The Indian populations of C. amata are also deeply C. eulimene (Klug 1829) divergent from their African counterparts (6.6 ± 0.4%), and (5) Genus Nepheronia Butler, 1870 morphologically distinct (i.e. wider wings and lighter salmon N. argia (Fabricius 1775) coloration). We did not examine the amata populations from N. buquetii (Boisduval, 1836) Arabia to Senegal, which may prove to have some interme- N. pauliani Bernardi, 1959 (stat. nov.) diate status. However, on the basis of the available data we N. pharis (Boisduval, 1836) reinstate all non-Indian amata to C. calais (stat. rev.) and N. thalassina (Boisduval, 1836) recognize the subspecies C. calais williami (Namibia and (6) Genus Pareronia Bingham, 1907 Angola) and C. calais crowleyi (Madagascar).
P. argolis (C. Felder and R. Felder, 1860)* (7) Colotis danae from India (ssp. danae) and Iran (ssp. dulcis) P. avatar (Moore, 1858)* are divergent from African populations (3.7 ± 0.9% and P. aviena Fruhstorfer, 1910* 3.9 ± 0.3% respectively) and from one another (3.4 ± 0.2%), P. boebera (Eschscholtz, 1821)* while only minor variation exists within the African C. danae P. ceylanica (C. Felder and R. Felder, 1865)* (0.5 ± 0.2%). We recognize the subspecies C. danae danae, P. chinki Joicey and Noakes, 1915* C. danae dulcis and C. danae eupompe (=pseudocaste syn.
P. anais Lesson, 1837 (=P. hippia (Fabricius, 1787)) nov.). The larger taxon annae Wallengren, 1857 ﬂies in dry P. iobaea (Boisduval, 1832)* parts of South Africa, from Natal to Namibia and north to P. nishiyamai Yata, 1981* Zambia and Botswana. Colotis hildebrandti Staudinger (1884) P. paravatar Bingham, 1907* is similarly large and ranges from southern Kenya to central P. phocaea (C. Felder and R. Felder, 1861)* Zambia. Their barcodes were similar (0.38 ± 0.05%). We P. tritaea (C. Felder and R. Felder, 1859)* reinstate C. annae (stat. rev.) as a good species, with ssp.
P. valeria (Cramer, 1776) hildebrandti as its northern subspecies (stat. nov.).
(7) Genus Eronia Hu¨bner, (1823) (8) We treat C. phisadia and C. vestalis as a single species, with E. cleodora Hu¨bner, (1823) an Arabian ⁄ African subspecies (i.e. C. phisadia phisadia) and E. dilatata Butler, 1888 (stat. rev.) an Asian subspecies (i.e. C. phisadia vestalis stat. nov.). We also (8) Genus Afrodryas Stoneham (1957) (stat. rev.) reinstate C. castalis (stat. rev.) as a separate species, described A. leda (Boisduval 1847) (comb. rev.) from Tanzania and found in dry habitats through Kenya to (9) Genus Pinacopteryx Wallengren (1857) P. eriphia (Godart, 1819) J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH Phylogenetic systematics of Colotis P. mabillei (Aurivillius, 1899) (stat. nov.) (10) Genus Teracolus Swainson, (1833) (stat. rev.) C. zoe (Grandidier, 1867) T. agoye (Wallengren 1857) (comb. rev.) C. celimene (Lucas, 1852) T. eris (Klug 1829) (comb. rev.) C. praeclarus (Butler, 1886) T. subfasciatus Swainson (1833) (comb. rev.) (11) Genus Colotis Hu¨bner, (1819) C. ungemachi (Le Cerf, 1922) C. doubledayi (Hopﬀer, 1862) C. etrida (Boisduval, 1836) C. chrysonome (Klug 1829) C. ephyia (Klug 1829) C. vesta (Reiche, 1850) C. aurigineus (Butler, 1883) C. aurora (Cramer, 1780) C. evarne (Klug 1829) (stat. rev.) C. protomedia (Klug 1829) C. incretus (Butler, 1881) (stat. rev.) C. halimede (Klug 1829) C. auxo (Lucas, 1852) C. pleione (Klug 1829) ssp. dissociatus (Butler, 1897) (stat. nov.) C. venosa (Staudinger 1884) C. mananhari (Ward, 1870) C. antevippe (Boisduval, 1836) C. regina (Trimen, 1863) C. rogersi (Dixey, 1915) C. hetaera (Gerstaecker, 1871) C. euippe (Linnaeus, 1758) C. elgonensis (Sharpe, 1891) C. pallene (Hopﬀer, 1855) C. ione (Godart, 1819) C. lais (Butler 1876) C. erone (Angas, 1849) C. daira (Klug 1829) C. eunoma (Hopﬀer, 1855)* C. evagore (Klug 1829)C. evanthe (Boisduval, 1836) ssp. evanthides (Holland, 1896) (stat. nov.) Supporting Information Additional Supporting Information may be found in the online C. liagore (Klug 1829) version of this article: C. evenina (Wallengren 1857) Figure S1. TCS haplotype networks for cytochrome oxidase I barcodes in the Colotis vesta (a) and Colotis phisadia (b) C. annae (Wallengren 1857) (stat. rev.) groups of species.
ssp. hildebrandti (Staudinger 1884) (stat. nov.) Table S1. Material examined and GenBank Accession C. guenei (Mabille, 1877) C. danae (Fabricius 1775) Table S2. Taxa and sequences used in phylogenetic recon- ssp. eupompe (Klug 1829) (=pseudacaste (Butler 1876) Table S3. Summary of character partitions.
ssp. dulcis (Butler 1876) Table S4. Support values for clades supported through both maximum likelihood and Bayesian analyses.
C. protractus (Butler 1876) Please note: Wiley-Blackwell are not responsible for the C. fausta (Olivier, 1804) content or functionality of any supporting materials supplied C. amata (Fabricius 1775) by the authors. Any queries (other than missing material) C. calais (Cramer, 1775) (stat. rev.) should be directed to the corresponding author for the article.
C. castalis (Staudinger, 1885) (stat. rev.)C. phisadia (Godart, 1819) ssp. vestalis (Butler 1876) (stat. nov.) J Zool Syst Evol Res (2011) 49(3), 204–215 2011 Blackwell Verlag GmbH
Antibiotic resistAnce profiles of EschErichia coli isolAted from different wAter sources in the mmAbAtho locAlity, north-west province, south AfricA Authors: Constance N. Wose Kinge1 The antibiotic resistance profiles of Escherichia coli (E. coli), isolated from different water sources in
Int.J.Curr.Microbiol.App.Sci (2014) 3(9) 573-581 ISSN: 2319-7706 Volume 3 Number 9 (2014) pp. 573-581 Original Research Article Studies on the comparison of phytochemical constituents and antimicrobial activity of Curcuma longa varieties S.Shanmugam* and P.Bhavani