Phylogenetic utility of elongation factor-1 alpha in noctuoidea (Insecta: Lepidoptera): the limits of synonymous substitution

To test its phylogenetic utility, nucleotide sequence variation in a 1,240-bp fragment of the elongation factor-1 alpha (EF-1 alpha) gene was examined in 49 moth species representing the major groups of the superfamily Noctuoidea. Both parsimony and distance analyses supported the monophyly of nearly all groups for which there are clear morphological synapomorphies. Clades of subfamily rank and lower, probably mid-Tertiary and younger, were strongly supported. The third codon position contains 88% of variable sites, and approaches saturation at approximately 20% sequence divergence, possibly due to among-site rate heterogeneity and composition bias; higher divergences occur only in association with shifts in composition. Surprisingly, the few nonsynonymous changes appear no more phylogenetically reliable than synonymous changes. Signal strength for basal divergences is weak and fails to improve with character weighting; thus, dense taxon sampling is probably needed for strong inference from EF-1 alpha regarding deeper splits in Noctuoidea (probably early Tertiary). EF-1 alpha synonymous changes show promise for phylogeny reconstruction within Noctuidae and other groups of Tertiary age.      

Evolution and phylogenetic utility of the period gene in Lepidoptera

Evolution and phylogenetic utility of the period gene are explored through sequence analysis of a relatively conserved 909-bp fragment in 26 lepidopteran species. Taxa range from tribes to superfamilies, primarily within the putative clade Macrolepidotera plus near outgroups, and include both strongly established and problematic groupings. Their divergence dates probably range from the late Cretaceous through much of the Tertiary. Comparisons within the same set of closely related species show that amino acid substitutions in period occur 4.9 and 44 times as frequently as they do in two other nuclear genes--dopa decarboxylase and elongation factor-1 alpha, respectively. In contrast, rates of observed synonymous substitution are within 60% of each other for these three genes. Synonymous changes in period approach saturation by the family level, whereas nonsynonymous and amino acid divergences across the Macrolepidoptera are less than half the maximal values reported for this gene. Phylogenetic analyses of period strongly supported groupings at the family level and below. In contrast to previous analyses at this level with other nuclear genes, much of the information lies in nonsynonymous change. Relationships up to the superfamily level were recovered with decreasing effectiveness, and little, if any, signal was apparent regarding relationships among superfamilies. This could reflect rapid radiation of the superfamilies, however, rather than saturation in the period locus; thus, period, in combination with other genes, remains a plausible candidate for approaching the difficult problems of lepidopteran family and superfamily relationships.      

Combined molecular and morphological evidence on the phylogeny of the earliest lepidopteran lineages

Agreement among recent morphological and molecular phylogenetic analyses has strengthened estimates of the relationships among the earliest lineages of the holometabolan order Lepidoptera. For a few major groups, evidence for monophyly and basal relationships remains relatively weak or contradictory — chiefly within the clades of basal Glossata and Heteroneura. Here we assess the support for these controversial areas of lepidopteran classification through molecular systematic investigation of 18S rDNA sequence variation. Parsimony and maximum likelihood analyses are presented for 1379 alignable sites of 18S. These data are then combined with 61 morphological features scored for major lineages of basal Glossata and Heteroneura. Our 18S rDNA data support recent hypotheses for the placement of Micropterigidae and Agathiphagidae as the basal-most lineages of Lepidoptera, and support the monophyly of the groups Neolepidoptera and Exoporia. 18S data alone are shown to be insufficient for resolving the monophyly and relationships of the Glossata, and for specifying relationships above the Neolepidoptera. Combination of the 18S data with published morphological ground-plan scorings improves overall support for the morphology-based hypothesis for basal glossatans, but phylogenetic resolution among published alternatives for the basal Heteroneura remains a major question for lepidopteran systematics.     

Deep sympatric mtDNA divergence in the autumnal moth (Epirrita autumnata)

Deep sympatric intraspecific divergence in mtDNA may reflect cryptic species or formerly distinct lineages in the process of remerging. Preliminary results from DNA barcoding of Scandinavian butterflies and moths showed high intraspecific sequence variation in the autumnal moth, Epirrita autumnata. In this study, specimens from different localities in Norway and some samples from Finland and Scotland, with two congeneric species as outgroups, were sequenced with mitochondrial and nuclear markers to resolve the discrepancy found between mtDNA divergence and present species‐level taxonomy. We found five COI sub‐clades within the E. autumnata complex, most of which were sympatric and with little geographic structure. Nuclear markers (ITS2 and Wingless) showed little variation and gave no indications that E. autumnata comprises more than one species. The samples were screened with primers for Wolbachia outer surface gene (wsp) and 12% of the samples tested positive. Two Wolbachia strains were associated with different mtDNA sub‐clades within E. autumnata, which may indicate indirect selection/selective sweeps on haplotypes. Our results demonstrate that deep mtDNA divergences are not synonymous with cryptic speciation and this has important implications for the use of mtDNA in species delimitation, like in DNA barcoding.     

Molecular diversity of Amansieae (Ceramiales,Rhodophyta) from the Hawaiian Islands: a multi‐marker assessment reveals high diversity within Amansia glomerata

Sixty‐one Hawaiian algal specimens corresponding to members of the tribe Amansieae (Amansia and Osmundaria) were compared through DNA sequence analysis. Short DNA barcode‐like sequences of mitochondrial cytochrome c oxidase subunit I (COI) and universal plastid amplicon (UPA) markers were obtained for as many of the specimens as possible, and a subset of specimens was also used for amplification and sequencing of the nuclear small‐subunit rRNA (SSU) gene for phylogenetic inference in a broader taxonomic context. Statistical parsimony analysis of the COI and UPA markers for A. glomerata produced relationships among the samples that were largely congruent with each other, although the UPA marker was more conserved. The COI marker yielded three lineages, and nucleotide divergences for these three lineages were intermediate to those typically reported for intraspecific and interspecific comparisons, suggesting that they represent either incipient species or a complex of closely related species. The COI and UPA sequences demonstrated little to no divergence for Osmundaria obtusiloba and the taxon referred to as Amansia fimbrifolia. In contrast, specimens identified as A. daemelii were identical in sequence to lineage 3 sequences of A. glomerata, and it is recommended that this taxon no longer be included in species lists for the Hawaiian flora. Phylogenetic reconstruction based on the SSU gene was largely unresolved, indicating that this marker may be of limited utility for this purpose in this group of algae, but a small amount of nucleotide variation was found for samples of A. glomerata.     

Synonymous substitution-rate constants in Escherichia coli and Salmonella typhimurium and their relationship to gene expression and selection pressure

Based on the differences in synonymous codon use between E. coli and S. typhimurium, the synonymous substitution rates can be estimated. In contrast to previous studies on the substitution rates in these two organisms, we use a kinetic model that explicitly takes the selection bias into account. The selection pressure on synonymous codons for a particular amino acid can be calculated from the observed codon bias. This offers a unique opportunity to study systematically the relationship between substitution-rate constants and selection pressure. The results indicate that the codon bias in these organisms is determined by a mutation-selection balance rather than by stabilizing selection. A best fit to the data implies that the mutation rate constant increases about threefold in genes at low expression levels relative to those that are highly expressed.Correspondence to: O.G. Berg     

Phylogenetic analysis of the genus Thricops Rondani (Diptera: Muscidae) based on molecular and morphological characters

Abstract. The muscid genus Thricops Rondani comprises forty‐four species and two subspecies restricted to the northern hemisphere. A species‐level phylogenetic analysis of Thricops was conducted using forty‐four morphological characters, 426 bp of the nuclear gene white and 523 bp spanning the 5′ end of the cytochrome c oxidase subunit I (COI), the tRNA leucine gene (L2 region) and the 3′ end of the cytochrome c oxidase subunit II (COII). Thirty‐nine species and two subspecies of Thricops were included in the analysis. Two species of Azelia Robineau‐Desvoidy and one species of Hydrotaea Robineau‐Desvoidy were used as outgroups. Morphological characters were coded for all included species, the mitochondrial gene fragment (COI + II) was sequenced for a subset of seventeen species of Thricops and three outgroup species, and white for twelve of those seventeen Thricops species and two outgroup species. Six separate maximum parsimony analyses were performed on three taxon sets of different sizes (n = 14, n = 20, n = 44). Results from the partition homogeneity test indicated no significant incongruence between data partitions, and four combined maximum parsimony analyses were conducted (DNA + morphology for n = 14; COI + II + morphology for n = 20; DNA + morphology for n = 20; DNA + morphology for n = 44). The relative contribution of each data partition to individual nodes was assessed using partitioned Bremer support. Strict consensus trees resulting from the unweighted analyses of each dataset are presented. Combination of datasets increased resolution for the small taxon set (n = 14), but not for the larger ones (n = 20, n = 44), most probably due to increasing amounts of missing data in the larger taxon sets. Results from both individual and combined analyses of the smaller taxon sets (n = 14, n = 20) provided support for the monophyly of Thricops and a complete division of the genus into two monophyletic subgroups. The strict consensus cladograms resulting from the analysis of the morphological data alone and the combined data for the large taxa set (n = 44) both supported the monophyly of the genus, but placed the species Thricops foveolatus (Zetterstedt) and Thricops bukowskii (Ringdahl) at the base of the ingroup, in a polytomy with a relatively well‐resolved branch comprising all remaining species of the genus. The basal position of these two species, included in the morphological taxon set but absent in the others, illustrates the potential pitfalls of taxon sampling and missing data in phylogenetic analyses. The synonymy of Alloeostylus with Thricops as proposed by previous authors was supported by our results. Relative contributions of different data partitions is discussed, with the mitochondrial sequence generally providing finer resolution and better branch support than white.     

Evaluation of DNA barcoding and identification of new haplomorphs in Canadian deerflies and horseflies

This paper reports the first tests of the suitability of the standardized mitochondrial cytochrome c oxidase subunit I (COI) barcoding system for the identification of Canadian deerflies and horseflies. Two additional mitochondrial molecular markers were used to determine whether unambiguous species recognition in tabanids can be achieved. Our 332 Canadian tabanid samples yielded 650 sequences from five genera and 42 species. Standard COI barcodes demonstrated a strong A + T bias (mean 68.1%), especially at third codon positions (mean 93.0%). Our preliminary test of this system showed that the standard COI barcode worked well for Canadian Tabanidae: the target DNA can be easily recovered from small amounts of insect tissue and aligned for all tabanid taxa. Each tabanid species possessed distinctive sets of COI haplotypes which discriminated well among species. Average conspecific Kimura two‐parameter (K2P) divergence (0.49%) was 12 times lower than the average divergence within species. Both the neighbour‐joining and the Bayesian methods produced trees with identical monophyletic species groups. Two species, Chrysops dawsoni Philip and Chrysops montanus Osten Sacken (Diptera: Tabanidae), showed relatively deep intraspecific sequence divergences (～10 times the average) for all three mitochondrial gene regions analysed. We suggest provisional differentiation of Ch. montanus into two haplotypes, namely, Ch. montanus haplomorph 1 and Ch. montanus haplomorph 2, both defined by their molecular sequences and by newly discovered differences in structural features near their ocelli.     

Cryptic diversity in flathead fishes (Scorpaeniformes: Platycephalidae) across the Indo‐West Pacific uncovered by DNA barcoding

Identification of taxonomical units underpins most biological endeavours ranging from accurate biodiversity estimates to the effective management of sustainably harvested, protected or endangered species. Successful species identification is now frequently based on a combination of approaches including morphometrics and DNA markers. Sequencing of the mitochondrial COI gene is an established methodology with an international campaign directed at barcoding all fishes. We employed COI sequencing alongside traditional taxonomic identification methods and uncovered instances of deep intraspecific genetic divergences among flathead species. Sixty‐five operational taxonomic units (OTUs) were observed across the Indo‐West Pacific from just 48 currently recognized species. The most comprehensively sampled taxon, Platycephalus indicus, exhibited the highest levels of genetic diversity with eight lineages separated by up to 16.37% genetic distance. Our results clearly indicate a thorough reappraisal of the current taxonomy of P. indicus (and its three junior synonyms) is warranted in conjunction with detailed taxonomic work on the other additional Platycephalidae OTUs detected by DNA barcoding.     

Molecular phylogeny of rootworms and related galerucine beetles (Coleoptera: Chrysomelidae)

The Galerucinae (Coleoptera: Chrysomelidae) sensu stricto (true galerucines) comprise a large assemblage of diverse phytophagous beetles containing over 5000 described species. Together with their sister taxon, the flea beetles, which differ from true galerucines by having the hind femora usually modified for jumping, the Galerucinae sensu lato comprises over 13 000 described species and is the largest natural group within the Chrysomelidae. Unlike the flea beetles, for which robust hierarchical classification schemes have not been erected, an existing taxonomic structure exists for the true galerucines, based mostly on the works of the late John Wilcox. In the most recent taxonomic list of the Galerucinae sensu stricto, five tribes were established comprising 29 sections housing 488 genera. The majority of the diversity within these tribes is found within the tribe Luperini, in which two genera, Monolepta and Diabrotica, are known to contain over 500 described species. Here, we extend the work from previous phylogenetic studies of the Galerucinae by analysing four amplicons from three gene regions (18S and 28S rRNA; COI) representing 249 taxa, providing the largest phylogenetic analysis of this taxon to date. Using two seven‐state RNA models, we combine five maximum likelihood models (RNA + DNA for the rRNAs; three separate DNA models for the COI codon positions) for these partitions and analyse the data under likelihood using Bayesian inference. The results of these two analyses are compared with those from equally weighted parsimony. Instead of choosing the results from one optimality criterion over another, either based on statistical support, tree topology or philosophical predisposition, we elect to draw attention to the similar results produced by all three analyses, illustrating the robustness of the data to these different analytical methods. In general, the results from all three analyses are consistent with each other and previous molecular phylogenetic reconstructions for Galerucinae, except that increased taxon sampling for several groups, namely the tribes Hylaspini and Oidini, has improved the phylogenetic position of these taxa. As with previous analyses, under‐sampled taxa, such as the Old World Metacyclini and all sections of the subtribe Luperina, continue to be unstable, with the few taxa representing these groups fluctuating in their positions based on the implemented optimality criterion. Nonetheless, we report here the most comprehensive phylogenetic estimation for the Galerucinae to date.     

The deep divergences of neornithine birds: a phylogenetic analysis of morphological characters

Consensus is elusive regarding the phylogenetic relationships among neornithine (crown clade) birds. The ongoing debate over their deep divergences is despite recent increases in available molecular sequence data and the publication of several larger morphological data sets. In the present study, the phylogenetic relationships among 43 neornithine higher taxa are addressed using a data set of 148 osteological and soft tissue characters, which is one of the largest to date. The Mesozoic non‐neornithine birds Apsaravis, Hesperornis, and Ichthyornis are used as outgroup taxa for this analysis. Thus, for the first time, a broad array of morphological characters (including both cranial and postcranial characters) are analyzed for an ingroup densely sampling Neornithes, with crown clade outgroups used to polarize these characters. The strict consensus cladogram of two most parsimonious trees resultant from 1000 replicate heuristic searches (random stepwise addition, tree‐bisection‐reconnection) recovered several previously identified clades; the at‐one‐time contentious clades Galloanseres (waterfowl, fowl, and allies) and Palaeognathae were supported. Most notably, our analysis recovered monophyly of Neoaves, i.e., all neognathous birds to the exclusion of the Galloanseres, although this clade was weakly supported. The recently proposed sister taxon relationship between Steatornithidae (oilbird) and Trogonidae (trogons) was recovered. The traditional taxon “Falconiformes” (Cathartidae, Sagittariidae, Accipitridae, and Falconidae) was not found to be monophyletic, as Strigiformes (owls) are placed as the sister taxon of (Falconidae + Accipitridae). Monophyly of the traditional “Gruiformes” (cranes and allies) and ”Ciconiiformes” (storks and allies) was also not recovered. The primary analysis resulted in support for a sister group relationship between Gaviidae (loons) and Podicipedidae (grebes)—foot‐propelled diving birds that share many features of the pelvis and hind limb. Exclusion of Gaviidae and reanalysis of the data set, however, recovered the sister group relationship between Phoenicopteridae (flamingos) and grebes recently proposed from molecular sequence data.     

A new nuclear gene for insect phylogenetics: dopa decarboxylase is informative of relationships within Heliothinae (Lepidoptera: Noctuidae)

The lack of a readily accessible roster of nuclear genes informative at various taxonomic levels is a bottleneck for molecular systematics. In this report, we describe the first phylogenetic application of the sequence that encodes the enzyme dopa decarboxylase (DDC). For 14 test species within the noctuid moth subfamily Heliothinae that represent the previously best-supported groupings, a 690-bp fragment of DDC resolved relationships that are largely concordant with prior evidence from elongation factor-1 alpha (EF-1 alpha), morphology, and allozymes. Although both synonymous and nonsynonymous changes occur in DDC substantially more rapidly than they do in EF-1 alpha, DDC divergences within Heliothinae are below saturation at all codon positions. Analysis of DDC and EF-1 alpha in combination resulted in increased bootstrap support for several groupings. As a first estimate of previously unresolved relationships, DDC sequences were analyzed from 16 additional heliothines, for a total of 30 heliothine species plus outgroups. Previous relationships based on DDC were generally stable with increased taxon sampling, although a two- to eightfold downweighting of codon position 3 was required for complete concordance with the 14-species result. The weighted strict consensus trees were largely resolved and were congruent with most although not all previous hypotheses based on either morphology or EF-1 alpha. The proposed phylogeny suggests that the major agricultural pest heliothines belong to a single clade, characterized by polyphagy and associated life history traits, within this largely host-specific moth subfamily. DDC holds much promise for phylogenetic analysis of Tertiary-age animal groups.     

Molecular Characterization of the 3'-Terminal Region of Turnip mosaic virus Isolates from Eastern China

Turnip mosaic virus (TuMV; genus Potyvirus, family Potyviridae) causes great losses to cruciferous crop production worldwide. The 3′‐terminal genomic sequences of eight TuMV isolates from eastern China were compared with those of 74 other Chinese TuMV isolates of known host origin in the GenBank and isolated during the past 25 years. The reported sequences of the eight TuMV isolates are 1125 or 1126‐nucleotides (nt) long excluding the poly(A) tail. They all contain one partial open reading frame of 912 nt, encoding 304 amino acids, followed by a stop codon and a non‐translated region of 209–210 nt. Results of phylogenetic analyses showed that Chinese TuMV isolates clustered into three groups: basal‐BR, Asian‐BR and world‐B. The ratios of non‐synonymous and synonymous substitutions and results of amino acid alignment provided evidence for purifying or negative selection in TuMV populations of China.     

Selective modes determine evolutionary rates,gene compactness and expression patterns in Brassica

It has been well documented that most nuclear protein‐coding genes in organisms can be classified into two categories: positively selected genes (PSGs) and negatively selected genes (NSGs). The characteristics and evolutionary fates of different types of genes, however, have been poorly understood. In this study, the rates of nonsynonymous substitution (K_a) and the rates of synonymous substitution (K_s) were investigated by comparing the orthologs between the two sequenced Brassica species, Brassica rapa and Brassica oleracea, and the evolutionary rates, gene structures, expression patterns, and codon bias were compared between PSGs and NSGs. The resulting data show that PSGs have higher protein evolutionary rates, lower synonymous substitution rates, shorter gene length, fewer exons, higher functional specificity, lower expression level, higher tissue‐specific expression and stronger codon bias than NSGs. Although the quantities and values are different, the relative features of PSGs and NSGs have been largely verified in the model species Arabidopsis. These data suggest that PSGs and NSGs differ not only under selective pressure (K_a/K_s), but also in their evolutionary, structural and functional properties, indicating that selective modes may serve as a determinant factor for measuring evolutionary rates, gene compactness and expression patterns in Brassica.     

Phylogeny of the great cats (Felidae: Pantherinae), and the influence of fossil taxa and missing characters

Molecular phylogenetic studies of the extant Pantherinae have resulted in a variety of different hypotheses of relationships. This study presents the results of a cladistic study encompassing 45 osteological and dental characters in the skull and mandible, as well as 13 soft‐tissue and behavioural characters. Analyzing extant pantherines with osteological data only resulted in two equally parsimonious trees, which differed only with respects to the jaguar, a taxon which shows morphological affinity to the tiger as well as the lion + leopard. Addition of soft‐tissue characters resolved this ambiguity, and led to markedly improved bootstrap values. The inclusion of fossil taxa did not have an impact on topology, but was important for a correct understanding of character evolution, due to the fossils having a combination of characters unlike those of any extant taxon. The clouded leopard is the most basal pantherine, followed by the snow leopard. The large pantherines are a well supported group, to which the snow leopard does not belong, contrary to some molecular studies. Panthera palaeosinensis is no tiger, but may be close to the stem group from which the tiger evolved. P. atrox and P. spelaea are not on the lion lineage, as traditionally assumed, but are successive outgroups to the lion + leopard, although the position of P. spelaea is tentative, but is supported by other lines of evidence such as brain anatomy. © The Willi Hennig Society 2008.     

Molecular considerations in the evolution of bacterial genes

Summary Synonymous and nonsynonymous substitution rates at the loci encoding glyceraldehyde-3-phosphate dehydrogenase (gap) and outer membrane protein 3A (ompA) were examined in 12 species of enteric bacteria. By examining homologous sequences in species of varying degrees of relatedness and of known phylogenetic relationships, we analyzed the patterns of synonymous and nonsynonymous substitutions within and among these genes. Although both loci accumulate synonymous substitutions at reduced rates due to codon usage bias, portions of thegap andompA reading frames show significant deviation in synonymous substitution rates not attributable to local codon bias. A paucity of synonymous substitutions in portions of theompA gene may reflect selection for a novel mRNA secondary structure. In addition, these studies allow comparisons of homologous protein-coding sequences (gap) in plants, animals, and bacteria, revealing differences in evolutionary constraints on this glycolytic enzyme in these lineages.     

The Codon-Degeneracy Model of Molecular Evolution

Mitochondrial genetic codons can be categorized by four patterns of nucleotide-site degeneracy based on varying combinations of twofold- or nondegenerate sites at first codon positions and twofold- or fourfold-degenerate sites at third codon positions. Herein, a model of molecular evolution is introduced that uses these patterns to calculate expected substitution frequencies for each codon position and substitution type relative to overall number of synonymous or nonsynonymous substitutions. Regions of the pocket gopher cytochrome oxidase subunit I (COI) and cytochrome b (cyt-b) genes are analyzed using this model. Chi-square distributions are used to produce relative goodness-of-fit (GF) scores for measuring the difference between substitution frequencies predicted by the codon-degeneracy model (CDM), and frequencies inferred using a well-supported phylogenetic tree of closely related species. The GF scores for expected and observed synonymous (GF_syn= 0.429, p= 0.807) and nonsynonymous (GF_ns= 2.309, p= 0.679) substitution frequencies resulted in a failure to reject the CDM as a null hypothesis for the molecular evolution of COI and cyt-b in pocket gophers. Alternative tree topologies and calculations of transition bias for these data result in higher GF scores. Received: 25 March 1999 / Accepted: 17 September 1999     

Deeply divergent lineages of the widespread New Zealand amphipod Paracalliope fluviatilis revealed using allozyme and mitochondrial DNA analyses

1. We evaluated the population genetic structure of the common New Zealand amphipod Paracalliope fluviatilis using eight allozyme loci, and the mitochondrial cytochrome oxidase c subunit I (COI) gene locus. Morphological analyses were also conducted to evaluate any phenotypic differences. Individuals belonging to P. fluviatilis were collected from a total of 14 freshwater fluvial habitats on the North and South Islands, New Zealand. 2. We found evidence for strong genetic differentiation among locations (Wright's F_ST > 0.25), and fixed differences (non‐shared alleles) at two of the eight allozyme loci indicating the possibility of previously unknown species. Analysis of a 545‐bp fragment of the COI locus was mostly congruent with the allozyme data and revealed the same deeply divergent lineages (sequence divergences up to 26%). 3. Clear genetic breaks were identified between North Island and South Island populations. North Island populations separated by <100 km also showed genetic differences between east and west draining watersheds (sequence divergence >12%). Accordingly, present‐day dispersal among hydrologically isolated habitats appears minimal for this taxon. 4. Although population differences were clearly shown by allozyme and mtDNA analyses, individuals were morphologically indistinguishable. This suggests that, as in North American and European taxa (e.g. Hyalella and Gammarus), morphological conservatism may be prevalent among New Zealand's freshwater amphipods. We conclude that molecular techniques, particularly the COI gene locus, may be powerful tools for resolving species that show no distinctive morphological differences.     

Detecting evolutionary rate heterogeneity among mangroves and their close terrestrial relatives

Mangroves form the dominant intertidal ecosystems and differ morphologically and physiologically from their close terrestrial relatives. We investigate the molecular evolutionary pattern of the typical mangrove family, i.e. Rhizophoraceae, and rate heterogeneity for the plastid matK and rbcL genes in different species of the family, as revealed by phylogenetic analyses and relative‐rate tests. Our study documents evolutionary rate heterogeneity in the Rhizophoraceae for the two genes: the mangrove genus Bruguiera has relatively slow substitution rates compared to the terrestrial genus Carallia at both synonymous and non‐synonymous sites in the matK sequences, and the synonymous and non‐synonymous substitution matrices are correlated. However, the rbcL non‐synonymous sites exhibit a high degree of rate heterogeneity among mangroves and related terrestrial groups, and uncoupling of rates with the synonymous sites. Selection is probably an important influence on the rate variation, suggesting further investigation for better understanding of various forces contributing to the rate heterogeneity and molecular adaptation in mangroves.