A molecular phylogeny of the tiger beetles (Cicindelidae): congruence of mitochondrial and nuclear rDNA data sets

Novelty and emergent patterns in sperm: Morphological diversity and evolution of spermatozoa and sperm conjugation in ground beetles (Coleoptera: Carabidae)

The beetle family Carabidae, with about 40,000 species, exhibits enough diversity in sperm structure and behavior to be an excellent model system for studying patterns and processes of evolution. We explore their potential, documenting sperm form in 177 species of ground beetles using light microscopy and collecting data on one qualitative and seven quantitative phenotypic traits. Our sampling captures 61% of the tribal-level diversity of ground beetles. These data highlight the notable morphological diversity of sperm in ground beetles and suggest that sperm in the group have dynamic evolutionary histories with much morphological innovation and convergence. Sperm vary among species in total length (48-3,400 μm), head length (0.5-270 μm), and head width (0.2-6.3 μm). Most ground beetles make sperm with heads that are indistinct from the flagella at the gross morphological level. However, some or all Omophron, Trachypachus, and Dyschiriini make broad-headed sperm that show morphological differences between species. Most ground beetles package their sperm into groups of sperm, termed conjugates, and ground beetles show variation in conjugate form and in the number and arrangement of sperm in a conjugate. Most ground beetles make sperm conjugates by embedding their sperm in a hyaline rod or spermatostyle. The spermatostyle is remarkably variable among species and varies in length from 17 to 41,000 μm. Several unrelated groups of ground beetles make only singleton sperm, including Nebriinae, Cicindelinae, many Trechinae, and the tribe Paussini. In order to study patterns in sperm evolution, we combine these data with a low-resolution phylogeny of ground beetles. Results from modern comparative analyses suggest the following: (a) sperm differ from conjugates in some aspect of their underlying evolutionary process, (b) sperm have influenced conjugate evolution and vice versa, and (c) conjugation with a spermatostyle likely evolved early within the history of Carabidae and it has been lost independently at least three times.

MitoFinder: Efficient automated large-scale extraction of mitogenomic data in target enrichment phylogenomics

Thanks to the development of high-throughput sequencing technologies, target enrichment sequencing of nuclear ultraconserved DNA elements (UCEs) now allows routine inference of phylogenetic relationships from thousands of genomic markers. Recently, it has been shown that mitochondrial DNA (mtDNA) is frequently sequenced alongside the targeted loci in such capture experiments. Despite its broad evolutionary interest, mtDNA is rarely assembled and used in conjunction with nuclear markers in capture-based studies. Here, we developed MitoFinder, a user-friendly bioinformatic pipeline, to efficiently assemble and annotate mitogenomic data from hundreds of UCE libraries. As a case study, we used ants (Formicidae) for which 501 UCE libraries have been sequenced whereas only 29 mitogenomes are available. We compared the efficiency of four different assemblers (IDBA-UD, MEGAHIT, MetaSPAdes, and Trinity) for assembling both UCE and mtDNA loci. Using MitoFinder, we show that metagenomic assemblers, in particular MetaSPAdes, are well suited to assemble both UCEs and mtDNA. Mitogenomic signal was successfully extracted from all 501 UCE libraries, allowing us to confirm species identification using CO1 barcoding. Moreover, our automated procedure retrieved 296 cases in which the mitochondrial genome was assembled in a single contig, thus increasing the number of available ant mitogenomes by an order of magnitude. By utilizing the power of metagenomic assemblers, MitoFinder provides an efficient tool to extract complementary mitogenomic data from UCE libraries, allowing testing for potential mitonuclear discordance. Our approach is potentially applicable to other sequence capture methods, transcriptomic data and whole genome shotgun sequencing in diverse taxa. The MitoFinder software is available from GitHub (https://github.com/RemiAllio/MitoFinder).

Transcriptomics illuminate the phylogenetic backbone of tiger beetles

Phylogenomics is progressing rapidly, allowing large strides forward into our understanding of the tree of life. In this study, we generated transcriptomes from ethanol-preserved specimens of 13 tiger beetle species (Coleoptera: Cicindelinae) and one Scaritinae outgroup. From these 14 transcriptomes and seven publicly available transcriptomes, we recovered an average of 2538 loci for phylogenetic analysis. We constructed an evolutionary tree of tiger beetles to examine deep-level relationships and examined the extent to which the composition of the dataset, missing data, gene tree inconsistency and codon position saturation impacted phylogenetic accuracy. Ethanol-preserved specimens yielded similar numbers of loci to specimens originally preserved in costly reagents, showcasing more flexibility in transcriptomics than anticipated. The number of loci and gene tree inconsistency had less impact on downstream results than third codon position saturation and missing data. Our results recovered tiger beetles as sister to Carabidae with strong support, confirming their taxonomic status as an independent family within Adephaga. Within tiger beetles, phylogenetic relationships were robust across all nodes. This new phylogenomic backbone represents a useful framework for future endeavours in tiger beetle systematics and serves as a starting point for the development of less costly target capture toolkits to expand the taxonomic breadth of the future tiger beetle tree of life.

New findings on the sperm ultrastructure of Carabidae (Insecta, Coleoptera)

The sperm structure of several species belonging to different tribes of the large Carabidae family is described. Some species of Nebriinae, such as Nebria brevicollis and Notiophilus biguttatus, have free conventional insect sperm. Their sperm type can be regarded as the ancestral model for Carabidae. All the other species examined, either with isolated sperm such as Calomera nemoralis, Scarites sp., Duvalius andreinii and Anillus florentinus or with spermatozeugmata and sperm associated to spermatostyles such as Typhloreicheia usslaubi, Brachinus italicus, Carabus convexus, Calathus fuscipes, Calathus montivagus, and Paraphorus mendax, showed sperm with long nucleus and a parallel axoneme running the length of the tail starting from the apical bell-like acrosome. C. nemoralis, like Cicindela campestris previously studied, has a sperm structure similar to that of several other Carabidae, confirming their correct assignment to the family. C. convexus has the same sperm structure as previously studied C. preslii and C. interstitialis, indicating that the spermatozeugmata of the group consist only of an apical cap in which the anterior sperm regions are embedded. Unlike other Carabidae with spermatozeugmata, Carabini have the typical sperm organization with acrosome, nucleus and flagellum in a regular sequence. A. florentinus, (Trechinae) shows major differences, such as the absence of an acrosome and an extremely long nucleus that reaches the end of the tail in close association with the axoneme. T. usslaubi (Scaritinae) has slender spermatozeugmata with orderly quartets of sperm. The posterior region of the sperm tail is also unusual, showing a perfect circular section and a plasma membrane reinforced by a dense underlying layer. The present observations confirm that spermatozeugmata, can vary in shape and size among different species of the Carabidae. Such diversity may be the result of the male reproductive strategy, different in each species, that enhances the efficiency of sperm transfer to the female.

Pleistocene phylogeography and cryptic diversity of a tiger beetle, Calomera littoralis, in North-Eastern Mediterranean and Pontic regions inferred from mitochondrial COI gene sequences

Background. Calomera littoralis is a Palearctic species, widely distributed in Europe; inhabiting predominantly its Atlantic, Mediterranean and Black Sea coastlines. Methods. Its phylogeography on the Balkan Peninsula and on the north-western Black Sea coast was inferred using a 697 bp long portion of the mitochondrial COI gene, amplified from 169 individuals collected on 43 localities. Results. The results revealed two genetically divergent groups/lineages, the southern one inhabiting both the Balkan Peninsula and the Pontic Region and the northern one found exclusively in the Pontic Region. Species delimitation based on DNA barcoding gap suggested an interspecific level of divergence between these groups. Multivariate analysis of eight male and female morphometric traits detected no difference between the groups, implying they may represent cryptic species. The Bayesian time-calibrated reconstruction of phylogeny suggested that the lineages diverged ca. 2.3 Ma, in early Pleistocene. Discussion. The presence of the two genetically divergent groups results most likely from contemporary isolation of the Pontic basin from the Mediterranean that broke the continuous strip of coastal habitats inhabited by C. littoralis. Demographic analyses indicated that both lineages have been in demographic and spatial expansion since ca. 0.15 Ma. It coincides with the terminal stage of MIS-6, i.e., Wartanian/Saalian glaciation, and beginning of MIS-5e, i.e., Eemian interglacial, during which, due to eustatic sea level rise, a wide connection between Mediterranean and the Pontic basin was re-established. This, along with re-appearance of coastal habitats could initiate north-east expansion of the southern lineage and its secondary contact with the northern one. The isolation of the Pontic basin from the Mediterranean during the Weichselian glaciation most likely did not have any effect on their phylogeography.

Static antennae act as locomotory guides that compensate for visual motion blur in a diurnal, keen-eyed predator

High visual acuity allows parallel processing of distant environmental features, but only when photons are abundant enough. Diurnal tiger beetles (Carabidae: Cicindelinae) have acute vision for insects and visually pursue prey in open, flat habitats. Their fast running speed causes motion blur that degrades visual contrast, forces stop-and-go pursuit and potentially impairs obstacle detection. We demonstrate here that vision is insufficient for obstacle detection during running, and show instead that antennal touch is both necessary and sufficient for obstacle detection. While running, tiger beetle vision appears to be photon-limited in a way reminiscent of animals in low-light habitats. Such animals often acquire wide-field spatial information through mechanosensation mediated by longer, more mobile appendages. We show that a nocturnal tiger beetle species waves its antennae in elliptical patterns typical of poorly sighted insects. While antennae of diurnal species are also used for mechanosensation, they are rigidly held forward with the tips close to the substrate. This enables timely detection of path obstructions followed by an increase in body pitch to avoid collision. Our results demonstrate adaptive mechanosensory augmentation of blurred visual information during fast locomotion, and suggest that future studies may reveal non-visual sensory compensation in other fast-moving animals.

Towards the phylogeny of chafers (Sericini): analysis of alignment-variable sequences and the evolution of segment numbers in the antennal club

Scarabaeoid beetles display a distinctive lamellate antenna carrying olfactory sensillae which show various trends of surface enlargement, including the increased number of the terminal lamellate antennomeres. The presence of >3 lamellae ('plurilamellate' antennae) in some groups has been used in the classification of chafers (Melolonthinae) and in particular in the tribe Sericini. However, this character may not be phylogenetically conservative. Here we present a phylogenetic analysis based on partial 28S rRNA, cytochrome oxidase I (cox1) and 16S rRNA (rrnL) for 183 species of Scarabaeidae, representing all traditionally recognized subfamilies, with particular focus on Sericini. Alignments of length-variable sequences were obtained applying various alignment algorithms and parameter settings. Tree topologies from the combined analysis were very similar when rrnL alignment was based on the progressive alignment algorithm MAFFT, MUSCLE, and less so Clustal, but differed greatly when using the probabilistic PRANK and the 'local' alignment procedure BlastAlign, while alignment conditions for the smaller 28S rRNA had little impact on the combined analysis. Preferred conditions were chosen based on an extensive analysis of character congruence between markers and recovery of well established taxonomic groups. Combined analyses on the best alignments using parsimony, maximum likelihood and Bayesian inference generally supported the traditional classification, including the monophyly of Scarabaeidae, with Glaphyridae as its sister, the monophyly of Cetoniinae, and the monophyly of most tribes included. Various levels of support were also obtained in favor of a proposed sister relationship of Sericini with Ablaberini, their close relationships to a melolonthine clade consisting of several tribes with exclusively Southern Hemisphere distribution, and the monophyly of Old World Sericini. In contrast, the generic level relationships were not consistent with the existing taxonomy. The large genera Neoserica, Microserica, and Maladera each split in several distantly related branches. The segment number of the antennal club when optimized onto the preferred tree revealed that plurilamellate antennae originated repeatedly (9-10 times in Sericini, plus multiple origins in other Melolonthinae). This invalidates the use of this trait in the generic classification. The number of lamellae is likely to be relevant to mate recognition, as it affects the spatial organization and number of olfactory sensillae. The high level of homoplasy in antennal characters may indicate a causal link between the morphological diversity of the antennae and the great species richness of Sericini and related melolonthines.

Resolution of the Hylobates phylogeny: congruence of mitochondrial D-loop sequences with molecular, behavioral, and morphological data sets

Gibbons of the genus Hylobates likely speciated very rapidly following isolation by rising sea levels during the Pleistocene. We sequenced the hypervariable region I (HV-I) of the mitochondrial D-loop to reconstruct the phylogeny of this group. Although the results clearly supported monophyly of each of the six species, the relationships among them were not clearly resolved by these data alone. A homogeneity test against published data sets of a coding mitochondrial locus (ND3-ND4 region), behavioral characters (vocalizations), and morphological traits (including skeletal and soft tissue anatomy) revealed no significant incongruence, and combining them resulted in a phylogenetic tree with much stronger support. The Kloss's gibbon (H. klossii), long considered a primitive taxon based on morphology, shares many molecular and vocal characteristics with the Javan gibbon (H. moloch), and appear as the most recently derived species. The northernmost species (H. lar and H. pileatus) are the most basal taxa. These data suggest that ancestral gibbons radiated from north to south. Unlike other markers, the HV-I region can accurately identify members of different gibbon species much like a DNA barcode, with potential applications to conservation.

On the constitution and phylogeny of Staphyliniformia (Insecta: Coleoptera)

The Staphyliniformia is one of the most diverse lineages of Coleoptera, with representatives occupying every conceivable non-marine niche. Phylogenetic relationships among its varied families and lower taxa have defied resolution. The problem has been further complicated by the recent suggestion that another major coleopteran series, Scarabaeiformia, is derived from within it. Here we present the first phylogenetic analyses, based on 18S rDNA sequences and morphological data, to explicitly examine this possibility. Thorough evaluation of alternative alignments and tree construction methods support the contention that Scarabaeiformia is derived from within Staphyliniformia. Though the analyses yielded strong support for few family level groupings within the expanded Staphyliniformia, they conclusively support a close relationship between Hydraenidae and Ptiliidae, which has often been debated. The primary factor hindering additional resolution appears to be the inconsistent rate of divergence in 18S among these taxa.

Cytogenetic analysis of Ctenostomini by C-banding and rDNA localization and its relevance to the knowledge of the evolution of tiger beetles (Coleoptera: Cicindelidae)

In this work, the first cytogenetic data on Neotropical Collyrinae is provided, by way of their karyotypes, C-banding and ribosomal genes (rDNA) localization using fluorescence in situ hybridization (FISH). The two species analysed, Ctenostoma (Procephalus) ornatum ornatum (male) and Ctenostoma (Euctenostoma) rugosum (female) showed, respectively, a diploid number of 17 and 18 chromosomes. C. ornatum ornatum has a multiple sex chromosome system (n = 7 + X1X2Y), and mitotic and meiotic metaphase cells showed rDNA gene labelling in the smallest autosomal pair. In this species, no C-bands were obtained, while C. rugosum seems to exhibit centromeric and/or interstitial C-bands in almost all chromosomes. The observation of a multiple sex chromosome system in Ctenostomini ensured the appearance of this characteristic in the hypothetical ancestral of Collyrinae and Cicindelini. The subfamily Collyrinae is not uniform in what concerns diploid chromosome number and rDNA gene localization, because C. ornatum ornatum possesses a lower chromosome number and autosomal rDNA genes when compared with the other Collyrinae species studied (Neocollyris spp.). Independent events leading to the reduction in chromosome number might have taken place during the split of the Collyrinae into the tribes Ctenostomini and Collyrini.

Independence of alignment and tree search

I assert that similarity is the appropriate homology criterion for sequence alignment, as it is with morphology. Methods that select among alignments using parsimony-based tree lengths, as implemented in MALIGN and POY, arrange the data such that they are consistent with a minimum-evolution model. When combining data sets in phylogenetic analyses, we are not trying to reinforce our earlier hypotheses about relationships, but rather to test them. The severity of this test is compromised when congruence with other characters is favored when selecting among alignment parameters.

Phylogeny of hydradephagan water beetles inferred from 18S rRNA sequences

Several families in the beetle suborder Adephaga have an aquatic life style and are commonly grouped in the "Hydradephaga," but their monophyly is contentious and relationships between and within these families are poorly understood. Here we present full-length 18S rRNA sequence for 84 species of Hydradephaga, including representatives of most major groups down to the tribal level, and a total of 68 species of the largest family, Dytiscidae. Using a direct optimization method for the alignment of length-variable regions, the preferred tree topology was obtained when the cost of gaps and the cost of nucleotide changes were equal, and three hypervariable regions of 18S rRNA were downweighted by a factor of five. Confirming recent molecular studies, the Hydradephaga were found to be monophyletic, indicating a single colonization of the aquatic medium. The most basal group within Hydradephaga is Gyrinidae, followed in a comb-like arrangement by families Haliplidae, Noteridae, Amphizoidae, and Hygrobiidae plus Dytiscidae. Under most alignment parameters, Hygrobiidae is placed amid Dytiscidae in an unstable position, suggesting a possible data artifact. Basal relationships within Dytiscidae are not well established, nor is the monophyly of subfamilies Hydroporinae and Colymbetinae. In contrast, relationships at the genus level appear generally well supported. Despite the great differences in the rates of change and the significant incongruence of the phylogenetic signal in conserved vs hypervariable regions of the 18S rRNA gene, both contribute to establish relationships at all taxonomic levels.

Secondary structure and conserved motifs of the frequently sequenced domains IV and V of the insect mitochondrial large subunit rRNA gene

We have analysed over 400 partial insect mitochondrial large subunit (mit LSU) sequences in order to identify conserved motifs and secondary structures for domains IV and V of this gene. Most of the secondary structure elements described by R. R. Gutell et al. (unpublished) for the LSU were identified. However, we present structures for helices 84 and 91 that are not recognized in previous universal models. The portion of the 16S gene containing domains IV and V is frequently sequenced in insect molecular systematic studies so we have many more sequences than previous studies which focused on the complete mitochondrial LSU molecule. In addition, we have the advantage of investigating several sets of closely related taxa. Aligned sequences from thirteen insect orders and nine secondary structure diagrams are presented. These conserved sequence motifs and their associated secondary structure elements can now be used to facilitate the alignment of other insect mit LSU sequences.

Phylogeny and biogeography of serolid isopods (Crustacea, Isopoda, Serolidae) and the use of ribosomal expansion segments in molecular systematics

In this study, a molecular phylogenetic hypothesis for 16 species of serolid isopods (Crustacea, Isopoda, Serolidae) from Antarctic waters, the deep sea, South America, and Australia is presented. The genes used are a 500-bp fragment of the mitochondrial LSU rRNA gene and a 700-bp fragment located in the variable region V4 of the nuclear SSU rRNA gene. The species composition and monophyly of morphologically defined genera of which several members were available are confirmed by the molecular data (Ceratoserolis, Spinoserolis, and Cuspidoserolis). The molecular data also support the redefinition of Frontoserolis s.l. and Serolella and the erection of the new genera Septemserolis and Paraserolis, as proposed by W]agele. The relationship among several genera is resolved differently in the molecular hypothesis than in the two existing morphological hypotheses, however. The molecular phylogeny may have important consequences for understanding the biogeography of the Serolidae, indicating that all Antarctic species in this study form a monophyletic group which has probably derived from species with closest extant relatives in South America. All 3 species included in this study living today in deep waters (>2000 m) of the Southern Ocean are most closely related to species living on the Antarctic shelf, so that parallel colonization of the deep sea by way of polar submergence can be reconstructed. In this study, a V4 expansion segment is reported which exceeds the longest crustacean sequences known until now by more than 270 bp. Although the V4 expansion segment has proven useful for phylogenetic purposes in this study, there is circumstantial evidence that its mechanism of evolution may depend not only on inheritance of single-site substitutions, making its routine use in phylogenetic studies potentially dangerous.

How slippage-derived sequences are incorporated into rRNA variable-region secondary structure: implications for phylogeny reconstruction

We analyzed the type and frequency of mutational changes in hypervariable rRNA regions, using the highly length-variable region V4 of the small subunit rRNA locus of tiger beetles (Cicindelidae) as an example. Phylogenetic analysis of indels in closely related species showed that (1) most indels are single nucleotides (usually A or T and sometimes G) or di-nucleotides of A and T. These occur at numerous foci, and they exhibit a strong bias for duplication of 5' single and di-nucleotide motifs but not 3' motifs. (2) Insertions/deletions in stem-forming regions affected paired and unpaired bases with about equal frequency but they did not disrupt the secondary structure. (3) Recurring mutations involving short repeats of the same bases caused parallel evolution of similar sequence motifs in the rRNA of different lineages. The observed types of change are consistent with the propostion that slippage is the main mutational mechanism. Slippage-derived sequences tend to be self-complementary, and therefore the stem-loop structure could be self-organizing as a consequence of the underlying mutational mechanism. Thus, the secondary structure in the cicindelid V4 region may be conserved due to the dynamics of the mutational mechanism rather than to functional constraints. These processes may also have a tendency to produce similar primary sequences irrespective of phylogenetic associations. The findings have implications for sequence alignment in phylogenetic analysis and should caution against the use of secondary structure to improve the determination of positional homology in hypervariable regions.

The current state of insect molecular systematics: a thriving Tower of Babel

Insect molecular systematics has undergone remarkable recent growth. Advances in methods of data generation and analysis have led to the accumulation of large amounts of DNA sequence data from most major insect groups. In addition to reviewing theoretical and methodological advances, we have compiled information on the taxa and regions sequenced from all available phylogenetic studies of insects. It is evident that investigators have not usually coordinated their efforts. The genes and regions that have been sequenced differ substantially among studies and the whole of our efforts is thus little greater than the sum of its parts. The cytochrome oxidase I, 16S, 18S, and elongation factor-1 alpha genes have been widely used and are informative across a broad range of divergences in insects. We advocate their use as standards for insect phylogenetics. Insect molecular systematics has complemented and enhanced the value of morphological and ecological data, making substantial contributions to evolutionary biology in the process. A more coordinated approach focused on gathering homologous sequence data will greatly facilitate such efforts.

Acetone preservation: a practical technique for molecular analysis

In attempts to establish a convenient and reliable method for field collection and archival preservation of insects and their endosymbiotic microorganisms for molecular analysis, acetone, ethanol, and other organic solvents were tested for DNA preservability of the pea aphid Acyrthosiphon pisum and its intracellular symbiotic bacterium Buchnera sp. After 6 months' storage, not only the band of high-molecular-size DNA but also the bands of rRNA were well preserved in acetone, ethanol, 2-propanol, diethyl ether and ethyl acetate. Polymerase chain reaction (PCR) assays confirmed that the DNA of both the insects and their symbionts was well preserved in these solvents. In contrast, methanol and chloroform showed poor DNA preservability. When water-containing series of acetone and ethanol were examined for DNA preservability, acetone was apparently more robust against water contamination than ethanol. Considering that most biological materials contain high amounts of water, acetone may be a more recommendable preservative for DNA analysis than ethanol which has been widely used for this purpose. The DNA of various insects could be preserved in acetone at room temperature in good condition for several years. In addition to the DNA of the host insects, the DNA of their endosymbionts, including Buchnera and other mycetocyte symbionts, Wolbachia, and gut bacteria, was amplified by PCR after several years of acetone storage. The RNA and protein of the pea aphid and its endosymbiont were also preserved for several years in acetone. After 2 years' storage in acetone, proteins of A. pisum could be analysed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting, and the endosymbiotic bacteria were successfully detected by immunohistochemistry and in situ hybridization on the tissue sections.

Modelling the secondary structures of slippage-prone hypervariable RNA regions: the example of the tiger beetle 18S rRNA variable region V4

Variable regions within ribosomal RNAs frequently vary in length as a result of incorporating products of slippage. This makes constructing secondary structure models problematic because base homology is difficult or impossible to establish between species. Here, we model such a region by comparing the results of the MFOLD suboptimal folding algorithm for different species to identify conserved structures. Based on the reconstruction of base change on a phylogenetic tree of the species and comparison against null models of character change, we devise a statistical analysis to assess support of these structures from compensatory and semi-compensatory (i.e. G.C to G.U or A.U to G.U) mutations. As a model system we have used variable region V4 from cicindelid (tiger beetle) small subunit ribosomal RNAs (SSU rRNAs). This consists of a mixture of conserved and highly variable subregions and has been subject to extensive comparative analysis in the past. The model that results is similar to a previously described model of this variable region derived from a different set of species and contains a novel structure in the central, highly variable part. The method we describe may be useful in modelling other RNA regions that are subject to slippage.

Phylogeny of North American Cicindela tiger beetles inferred from multiple mitochondrial DNA sequences

Tiger beetles in the genus Cicindela (Coleoptera: Cicindelidae) have been used as a model system for studies in ecology and conservation biology. Work on this group will greatly benefit from the availability of a phylogenetic hypothesis. We selected a representative sample of 23 North American Cicindela and 6 outgroups to reconstruct a phylogeny based on 1896 nucleotide positions from three mitochondrial genes (Cytochrome b, Cytochrome oxidase III, and 16S rRNA). Cladistic analysis of these three data sets yielded widely different tree topologies, but character conflict between them appears to be relatively low. The combined analysis of all data resulted in three similar shortest trees of 3453 steps. One of these was also recovered after successive weighting and was considered the best estimate of relationships. The most basal taxa of North American Cicindela (s.l.) were in the cosmopolitan subgenus Cylindera. The derived taxa were in the subgenus Cicindela (s. str.), a group dominating at higher latitudes in the Nearctic and Palearctic Region. The molecular analysis was essentially in agreement with the traditional classification which has been worked out based on male genitalic structures by E. Rivalier (1954, Rev. Entomol. Française 21:249-268). In the molecular analysis, Rivalier's species groups and subgenera were mostly found to be composed of closely related taxa but several of them were not monophyletic. Implicit in the traditional classification is a sequence from basal to derived groups which we found to be essentially reversed in the molecular analysis. We also discuss the conceptual differences in the establishment of the traditional classification by Rivalier (1954) and the cladistic analysis presented in this study.