Abstract.---Phylogenetic relationships among the holometabolous insect orders were inferred from cladistic analysis of nucleotide sequences of 18S ribosomal DNA (rDNA) (85 exemplars) and 28S rDNA (52 exemplars) and morphological characters. Exemplar outgroup taxa were Collembola (1 sequence), Archaeognatha (1), Ephemerida (1), Odonata (2), Plecoptera (2), Blattodea (1), Mantodea (1), Dermaptera (1), Orthoptera (1), Phasmatodea (1), Embioptera (1), Psocoptera (1), Phthiraptera (1), Hemiptera (4), and Thysanoptera (1). Exemplar ingroup taxa were Coleoptera: Archostemata (1), Adephaga (2), and Polyphaga (7); Megaloptera (1); Raphidioptera (1); Neuroptera (sensu stricto ;eq Planipennia): Mantispoidea (2), Hemerobioidea (2), and Myrmeleontoidea (2); Hymenoptera: Symphyta (4) and Apocrita (19); Trichoptera: Hydropsychoidea (1) and Limnephiloidea (2); Lepidoptera: Ditrysia (3); Siphonaptera: Pulicoidea (1) and Ceratophylloidea (2); Mecoptera: Meropeidae (1), Boreidae (1), Panorpidae (1), and Bittacidae (2); Diptera: Nematocera (1), Brachycera (2), and Cyclorrhapha (1); and Strepsiptera: Corioxenidae (1), Myrmecolacidae (1), Elenchidae (1), and Stylopidae (3). We analyzed ~1 kilobase of 18S rDNA, starting 398 nucleotides downstream of the 5' end, and ~400 bp of 28S rDNA in expansion segment D3. Multiple alignment of the 18S and 28S sequences resulted in 1,116 nucleotide positions with 24 insert regions and 398 positions with 14 insert regions, respectively. All Strepsiptera and Neuroptera have large insert regions in 18S and 28S. The secondary structure of 18S insert 23 is composed of long stems that are GC rich in the basal Strepsiptera and AT rich in the more derived Strepsiptera. A matrix of 176 morphological characters was analyzed for holometabolous orders. Incongruence length difference tests indicate that the 28S + morphological data sets are incongruent but that 28S + 18S, 18S + morphology, and 28S + 18S + morphology fail to reject the hypothesis of congruence. Phylogenetic trees were generated by parsimony analysis, and clade robustness was evaluated by branch length, Bremer support, percentage of extra steps required to force paraphyly, and sensitivity analysis using the following parameters: gap weights, morphological character weights, methods of data set combination, removal of key taxa, and alignment region. The following are monophyletic under most or all combinations of parameter values: Holometabola, Polyphaga, Megaloptera + Raphidioptera, Neuroptera, Hymenoptera, Trichoptera, Lepidoptera, Amphiesmenoptera (Trichoptera + Lepidoptera), Siphonaptera, Siphonaptera + Mecoptera, Strepsiptera, Diptera, and Strepsiptera + Diptera (Halteria). Antliophora (Mecoptera + Diptera + Siphonaptera + Strepsiptera), Mecopterida (Antliophora + Amphiesmenoptera), and Hymenoptera + Mecopterida are supported in the majority of total evidence analyses. Mecoptera may be paraphyletic because Boreus is often placed as sister group to the fleas; hence, Siphonaptera may be subordinate within Mecoptera. The 18S sequences for Priacma (Coleoptera: Archostemata), Colpocaccus (Coleoptera: Adephaga), Agulla (Raphidioptera), and Corydalus (Megaloptera) are nearly identical, and Neuropterida are monophyletic only when those two beetle sequences are removed from the analysis. Coleoptera are therefore paraphyletic under almost all combinations of parameter values. Halteria and Amphiesmenoptera have high Bremer support values and long branch lengths. The data do not support placement of Strepsiptera outside of Holometabola nor as sister group to Coleoptera. We reject the notion that the monophyly of Halteria is due to long branch attraction because Strepsiptera and Diptera do not have the longest branches and there is phylogenetic congruence between molecules, across the entire parameter space, and between morphological and molecular data.
[Alignment; insect orders; molecular systematics; phylogeny; sensitivity analysis; Strepsiptera; taxon sampling]

Abstract.---Although long-branch attraction, the incorrect grouping of long lineages in a phylogeny because of systematic error, has been identified as a potential source of error in phylogenetic analysis for almost two decades, no empirical examples of the phenomenon exist. Here, I outline several criteria for identifying long-branch attraction and apply these criteria to 18S ribosomal DNA (rDNA) sequence data for 13 insects. Parsimony and minimum evolution with p distances group the two longest branches together (those leading to Strepsiptera and Diptera). Simulation studies show that the long branches are long enough to attract. When a tree is assumed in which Strepsiptera and Diptera are separated and many data sets are simulated for that tree (using the parameter estimates for that tree for the original data), parsimony analysis of the simulated data consistently groups Strepsiptera and Diptera. Analyses of the 18S rDNA sequences using methods that are less sensitive to the problem of long-branch attraction estimate trees in which the long branches are separate and thus that are consistent with results of morphological studies.
[Felsenstein zone; insect phylogeny; long-branch attraction; neighbor joining; parsimony.]

Abstract.---Phylogenetic relationships among fishes from ostariophysan orders, characiform families, and serrasalmin genera (e.g., "piranhas") were examined using partial mitochondrial ribosomal DNA sequences of the 12S and 16S genes. Phylogenetic information content of these sequences was assessed at three levels of taxonomic inclusiveness by analyzing the patterns of nucleotide substitution using secondary structure models. Conserved and variable regions were identified, mapped onto the structural models, and compared at increasing levels of taxonomic divergence. In general, loop regions (unpaired) exhibited a higher level of variation than did stem regions (paired). A high proportion of compensatory substitutions was observed in stem regions in three data sets, suggesting strong selection to maintain the secondary structure. Saturation due to multiple substitutions was indicated by decreasing transition;shtransversion ratios and strong structural constraints on variation in comparisons among orders of Ostariophysi but was not obvious among familes of Characiformes and was not detected among serrasalmin genera. Reliable phylogenetic signal successfully reconstructed relationships among serrasalmin genera. However, aside from a few well-supported clades, relationships could not be reconstructed with confidence among characiform families and ostariophysan orders. The reciprocal monophyly of African and Neotropical characiform lineages was rejected (based on maximum likelihood ratio tests), and some support for previous hypotheses based on morphology was provided by the molecular data. The radiation of characiform fishes is discussed in a historical biogeographic context.
[Biogeography; Characiformes; mitochondrial DNA; molecular phylogenetics; Ostariophysi; phylogenetic information; secondary structure.]

Abstract.---A computational method is presented for minimizing the weighted sum of squares of the differences between observed and expected pairwise distances between species, where the expectations are generated by an additive tree model. The criteria of Fitch and Margoliash (1967, Science 155:279--284) and Cavalli-Sforza and Edwards (1967, Evolution 21:550--570) are both weighted least squares, with different weights. The method presented iterates lengths of adjacent branches in the tree three at a time. The weighted sum of squares never increases during the process of iteration, and the iterates approach a stationary point on the surface of the sum of squares. This iterative approach makes it particularly easy to maintain the constraint that branch lengths never become negative, although negative branch lengths can also be allowed. The method is implemented in a computer program, FITCH, which has been distributed since 1982 as part of the PHYLIP package of programs for inferring phylogenies, and is also implemented in PAUP*. The present method is compared, using some simulated data sets, with an implementation of the method of De Soete (1983, Psychometrika 48:621--626); it is slower than De Soete's method but more effective at finding the least squares tree. The relationship of this method to the neighbor-joining method is also discussed.
[Alternating least squares; distance methods; Fitch--Margoliash method; phylogenies.]

Abstract.---An experimental study on the delimitation of character states in continuous variation indicates that (1) the way data are presented influences the assignment of character states and (2) states in the same data set are delimited in various ways by different individuals. Forty-nine individuals were given a set of graphs denoting variation of 10 characters in the genus Kalmia (Ericaceae) and outgroups, all identification having been removed from the graphs. The variation was represented in one of three ways: as 95% confidence intervals on a linear scale, as 95% confidence intervals on a log10 scale, or with bars showing SD ;ts 2 on a linear scale. No two individuals scored a set of graphs in the same way, and only one character in one representation was scored identically by all individuals; the scoring for this character was completely different when the ordinate was changed from linear to logarithmic. Together, the 49 individuals delimited states within each character between 9 and 16 different ways. In general, variation represented by 2 ;ts SD bars elicited the largest numbers of different scorings, yet with a relatively low number of states; the complexity of the patterns in the graphs in this representation was greatest. Expert knowledge appears to be of dubious value in delimiting states in such variation, and if such characters are to be used in phylogenetic analyses, states could be delimited by people who know nothing of the details of the study being scored; in any case, presentation of data and an explicit protocol to follow when delimiting states are essential. In converting data of this type into character states, psychological factors are particularly likely to come into play. Other implications of our experiments include the severe underdetermination of some phylogenetic hypotheses by observation and the heterogeneous nature of morphological data.
[Character states; cognition; continuous variation; homology; observer variation; similarity.]

Abstract.---Sequence divergence for segments of three mitochondrial DNA (mtDNA) genes encoding the 12S and 16S ribosomal RNA and cytochrome b was examined in newts belonging to the genus Euproctus (E. asper, E. montanus, E. platycephalus) and in three other species belonging to the same family (Salamandridae), Triturus carnifex, T. vulgaris, and Pleurodeles waltl. The three Euproctus species occur (one species each) in Corsica, Sardinia, and the Pyrenees. This vicariant distribution is believed to have been determined by the disjunction and rotation of the Sardinia--Corsica microplate from the Pyreneean region. Because time estimates are available for the tectonic events that led to the separation of the three landmasses, we used sequence data to estimate rates of evolution for the three gene fragments and investigated whether they conform to the rate-constancy hypothesis. By the Tajima (1993, Genetics 135:599--607) test, we could not detect rate heterogeneities for the ribosomal genes and for transversions in the cytochrome b gene. Assuming that these sites are evolving linearly over time and that cessation of gene flow occurred simultaneously with vicariant events, we compared the time of divergence estimated by molecular distances with the divergence times based on the geological estimates. Because we have two estimates of divergence time from the geological record, the split of Corsica;shSardinia from the Pyrenees and the split of Corsica from Sardinia, we could compare ratios of molecular divergence with the ratio of geological time divergence. The ratios are very similar, indicating that the molecular clock hypothesis cannot be rejected. These geological events also allowed us to calculate absolute rates of evolution for ribosomal and cytochrome b genes and compare them to rates for the same regions in other salamandrids and other vertebrates. Ribosomal mtDNA rates are comparable to those reported for other vertebrates, but cytochrome b rates are 3--7 times lower in salamanders than in other ectotherms. From a phylogenetic perspective, our data suggest that the cladogenic events leading to species formation in Euproctus and Triturus occurred very closely in time, indicating that the two genera may not be monophyletic. A duplication of the cytochrome b gene in T. carnifex was found, and the implications of this finding for mtDNA phylogenetic studies are discussed.
[Biogeography; cytochrome b; Euproctus; mitochondrial DNA rates; mitochondrial ribosomal genes; molecular clocks; Triturus.]

Abstract.---The evolutionary pathway that has led to male tails of diverse morphology among species of the nematode family Rhabditidae was reconstructed. This family includes the well-studied model species Caenorhabditis elegans. By relating the steps of male tail morphological evolution to the phenotypic changes brought about by developmental mutations induced experimentally in C. elegans, the goal is to identify genes responsible for morphological evolution. The varying morphological characters of the male tails of several rhabditid species have been described previously (Fitch and Emmons, 1995, Dev. Biol. 170:564--582). The developmental events preceding differentiation of the adult structures have also been analyzed; in many cases the origins of varying adult morphological characters were traced to differences during ontogeny. In the present work, the evolutionary changes producing these differences were reconstructed in the context of the four possible phylogenies supported independently by sequences of 18S ribosomal RNA genes (rDNA). Two or more alternative states were defined for 36 developmental and adult morphological characters. These characters alone do not provide sufficient data to resolve most species relationships; however, when combined with the rDNA characters, they provide stronger support for one of the four rDNA phylogenies. Assuming a model of ordered transformations for multistate developmental characters generally results in greater resolution. Transformations between character states can be assigned unequivocally by parsimony to unambiguous branches for most of the characters. Correlations are thereby revealed for some of the developmental characters, indicating a probability of a shared developmental or genetic regulatory pathway. Four of the unequivocal character state changes on unambiguously supported branches closely resemble the phenotypic changes brought about by known mutations in C. elegans. These mutations define genes that are known to act in genetic regulatory hierarchies controlling pattern formation, differentiation, and morphogenesis. Although these studies are still at an early stage, these results strongly suggest that parallel studies of developmental mutants in C. elegans and of morphological and developmental evolution among related nematodes will help define genetic changes underlying the evolution of form.
[Caenorhabditis elegans; character analysis; evolution of development; ontogeny; phylogeny.]

Abstract.---Nominally polyphagous species of herbivorous insects sometimes are comprised of multiple morphologically similar biological species with more specialized appetites. When meristic morphological traits cannot be found to distinguish such suspected sibling species, molecular data are increasingly sought as a source of evidence. A role for morphology in distinguishing such taxa might be reclaimed, however, by recent advances in geometric morphometric methods, such as the statistical analysis of partial-warp scores from the thin-plate spline. We employed this method to detect and characterize subtle shape differences among populations and between sexes of the nominal leaf beetle species Neochlamisus bebbianae. Using the thin-plate spline, the shapes of specimens from seven beetle populations collected from five host plants in five eastern North American localities were calculated. These shapes were analyzed by MANOVA, revealing significant variation in both uniform and nonuniform components of shape among test populations. Significant sexual dimorphism in size, shape, and allometric relationships were also documented across these populations. More interestingly, our study provided evidence of sibling species where traditional taxonomic approaches have failed. Individual MANOVAs revealed significant shape variation between sympatric populations from different host plants in each of three localities. Because these sympatric shape differences were significant when adjusted for size, they cannot be attributed to allometric consequences of size variation among test populations. Because certain beetle populations differed significantly in size and shape when reared in a common environment, these morphometric traits may have a genetic basis. Together, these results are consistent with an earlier suggestion that N. bebbianae represents a complex of host-specific races or sibling species, a hypothesis that has received additional support from recent studies on host use traits, sex ratios, and mitochondrial DNA. In sum, these analyses demonstrate the power and utility of the thin-plate spline as a morphological means of discriminating among closely related and anatomically homogeneous taxa.
[Chrysomelidae; host races; morphometrics; Neochlamisus; relative warps; sexual dimorphism; sibling species; thin-plate spline.]

Abstract.---Quantification in historical biogeography has usually been based on the search for a single branching relationship among areas of endemism. Unlike organisms, however, areas rarely have a unique hierarchical history. Dispersal barriers appear and disappear and may have different effects on different species. As a result, the biota of an area may consist of several components with separate histories, each of which may be reticulate rather than branching. In an attempt to address these problems, I present a new biogeographic method, dispersal--vicariance analysis, which reconstructs the ancestral distributions in a given phylogeny without any prior assumptions about the form of area relationships. A three-dimensional step matrix based on a simple biogeographic model is used in the optimization. Speciation is assumed to subdivide the ranges of widespread species into vicariant components; the optimal ancestral distributions are those that minimize the number of implied dispersal and extinction events. Exact algorithms that find the optimal reconstruction(s) are described. In addition to their use in taxon biogeography, the inferred distribution histories of individual groups serve as a basis for the study of general patterns in historical biogeography, particularly if the relative age of the nodes in the source cladograms is known.
[Cladistic biogeography; comparative phylogeography; dispersal; extinction; historical biogeography; optimization; vicariance; widespread species.]