Minimum Length Trees Research Articles

Delay-related secondary objectives for rectilinear Steiner minimum trees

The rectilinear Steiner tree problem in the plane is to construct a minimum-length tree interconnecting a set of points (called terminals) consisting of horizontal and vertical line segments only. Rectilinear Steiner minimum trees (RSMTs) can today be computed quickly for realistic instances occurring in VLSI design. However, interconnect signal delays are becoming increasingly important in modern chip designs. Therefore, the length of paths or direct delay measures should be taken into account when constructing rectilinear Steiner trees. We consider the problem of finding an RSMT that — as a secondary objective — minimizes a signal delay related objective. Given a source (one of the terminals) we give some structural properties of RSMTs for which the weighted sum of path lengths from the source to the other terminals is minimized. Also, we present exact and heuristic algorithms for constructing RSMTs with weighted sum of path lengths or Elmore delays secondary objectives. Computational results for industrial designs are presented.

The full Steiner tree problem

Motivated by the reconstruction of phylogenetic tree in biology, we study the full Steiner tree problem in this paper. Given a complete graph G=(V,E) with a length function on E and a proper subset R⊂V, the problem is to find a full Steiner tree of minimum length in G, which is a kind of Steiner tree with all the vertices of R as its leaves. In this paper, we show that this problem is NP-complete and MAX SNP-hard, even when the lengths of the edges are restricted to either 1 or 2. For the instances with lengths either 1 or 2, we give a 85-approximation algorithm to find an approximate solution for the problem.

The phylogeny of Lens (Leguminosae): new insight from ITS sequence analysis

Lens includes L.culinaris subsp. culinaris (the cultivated lentil) and several wild species distributed from the Mediterranean region to western Asia. We compared sequence variation in the ITS region among species of Lens in an effort to end persisting uncertainty regarding the phylogeny of the genus. The parsimony analysis revealed a single minimum-length tree with a topology congruent with patterns derived by previous studies of nuclear and chloroplast DNA RFLPs. The basal and highly divergent status of the L. nigricans clade is depicted, and the progenitor-derivative relationship between L. culinaris subsp. orientalis and L. culinaris subsp. culinaris is reaffirmed. Resolution in the tree was improved by combining the ITS data set with a pre-existing set of chloroplast DNA restriction site data obtained from the same group of samples.

The Epicopeiidae: phylogeny and a redefinition, with the description of new taxa (Lepidoptera: Drepanoidea)

Amana banghaasi Hering, a poorly known species from central China (described in the “Epiplemidae”, now Epipleminae [Uraniidae]), is transferred to a new genus, Deuveia, which belongs to the Epicopeiidae (Drepanoidea). This taxon turns out to be the sister-group of a clade consisting of all other members of the family. Another new epicopeiid genus, Burmeia, is proposed for B. leesi n. sp., a species described from northern Burma. Quite clearly, there is a sister-group relationship between Burmeia and Psychostrophia, but these genera differ markedly in certain characters, for example in the course of vein M2: in both pairs of wings, M2 arises distinctly closer to M3 than to M1 in Burmeia, unlike the condition found in all other Epicopeiidae. A key is provided for the identification of the species in these two genera. A manually derived cladogram may correctly sum up the phylogeny of the Epicopeiidae, a family now composed of nine genera. It matches exactly one of the two most parsimonious trees found with computer programs such as Hennig86 or PAUP (the second tree of minimal length being similar, except for the affinities of the monotypic genus Amana). For the software-based analyses, all the 34 (imaginal) characters taken into account were polarized and given equal weight. The “robustness” of each branch of the preferred cladogram was assessed by calculating the corresponding Bremer’s support index, but also with non-quantifiable criteria. Six genera may form a clade, within which the following sister-group relationships can be considered well established: Chatamla/Parabraxas, Nossa/Epicopeia, and Schistomitra/Nossa + Epicopeia (whereas Chatamla + Parabraxas is only tentatively regarded as sister to Amana). Owing to the basal position of Deuveia within the Epicopeiidae, the morphology of this genus is of great significance to identify the autapomorphies of the Drepanoidea and those of the Epicopeiidae (respectively 5 and 12 in number at this stage of the investigations). With regard to the Drepanidae, an apomorphy of the male genitalia can be added to the definition of the family, whose basalmost lineage probably is the subfamily Cyclidiinae. The composition of the Thyatirinae is briefly discussed.

Origin and relationships of the tarweed–silversword lineage (Compositae–Madiinae)

Based on results from phylogenetic analyses of nuclear 18S-26S rDNA internal transcribed spacer (ITS) region sequences, we suggest that the monophyletic tarweed and silversword subtribe (Madiinae) is phylogenetically nested among epaleate, x = 19 species of helenioid Heliantheae. Strong bootstrap support (100%) was obtained for a sister-group relationship between Madiinae and Arnica (including Mallotopus and Whitneya) in an analysis including representatives of recognized genera in a principally Californian clade (Madieae sensu Baldwin) identified from a phylogenetic investigation of Heliantheae s.l. (sensu lato) and Eupatorieae. In all minimum-length trees, the robust lineage comprising Madiinae and Arnica (x = 19) is part of a larger clade that also comprises Eatonella s.s. (sensu stricto), Hulsea, and Venegasia, all with x = 19. The phylogenetic position of Madiinae within a group of genera based uniformly on x = 19 leads us to conclude that the modal numbers of n = 7 and n = 8 (and other numbers, as low as n = 4) in Madiinae are the results of extreme dysploidy. Among the x = 19 "arnicoid" taxa, the near-universal characteristics of perenniality (except in the monotypic Eatonella s.s. and a minority of hulseas) and montane or high-latitudinal occurrence (except in the monotypic Venegasia) lead us to suggest that the most recent common ancestor of the tarweeds (a principally annual group of seasonally dry, low-elevation habitats) was probably a montane, herbaceous perennial resembling the unusual subalpine and alpine tarweeds constituting Raillardella s.s. (x = 17), an arnica-like genus. In Madiinae, Raillardella s.s. may be plesiomorphic in habit, capitular and ecological characteristics, and high base chromosome number. Shifts to an annual habit and to low chromosome numbers in Madiinae have been followed by subsequent episodes of polyploidy and descending dysploidy. We conclude that genome evolution in Madiinae has been marked by wide swings in chromosome number that confuse identification of diploids and polyploids.

Choosing among Alternative Trees of Multigene Families

Estimation of gene trees is the first step in testing alternative hypotheses about the evolution of multigene families. The standard practice for inferring gene family history is to construct trees that meet some objective criteria based on the fit of the character state changes (nucleotide or amino acid changes) to the gene tree. Unfortunately, analysis of character state data can be misleading. In addition, this approach ignores information about the relationships of the species from which the genes have been sampled. In this paper I explore using statistics of fit between the character data and gene trees and the reconciliation of the gene and species trees for choosing among alternative evolutionary hypotheses of gene families. In particular, I advocate a two-pronged strategy for choosing among alternative gene trees. First, the character data are used to define a set of acceptable gene trees (i.e., trees that are not significantly different from the minimum length tree). Next, the set of acceptable gene trees is reconciled with a known species tree, and the gene tree requiring the fewest number of gene duplications and losses is adopted as the best estimate of evolutionary history. The approach is illustrated using three gene families: BMP, EGR, and LDH.

The Namanereidinae (Polychaeta: Nereididae). Part 1. Taxonomy and phylogeny

A cladistic analysis and taxonomic revision of the Namanereidinae (Nereididae: Polychaeta) is presented. The cladistic analysis utilising 39 morphological characters (76 apomorphic states) yielded 10,000 minimal-length trees and a highly unresolved Strict Consensus tree. However, monophyly of the Namanereidinae is supported and two clades are identified: Namalycastis containing 18 species and Namanereis containing 15 species. The monospecific genus Lycastoides, represented by L. alticola Johnson, is too poorly known to be included in the analysis. Classification of the subfamily is modified to reflect the phylogeny. Thus, Namalycastis includes large-bodied species having four pairs of tentacular cirri; autapomorphies include the presence of short, subconical antennae and enlarged, flattened and leaf-like posterior cirrophores. Namanereis includes smaller-bodied species having three or four pairs of tentacular cirri; autapomorphies include the absence of dorsal cirrophores, absence of notosetae and a tripartite pygidium. Cryptonereis Gibbs, Lycastella Feuerborn, Lycastilla Solis-Weiss & Espinasa and Lycastopsis Augener become junior synonyms of Namanereis. Thirty-six species are described, including seven new species of Namalycastis (N. arista n.sp., N. borealis n.sp., N. elobeyensis n.sp., N. intermedia n.sp., N. macroplatis n.sp., N. multiseta n.sp., N. nicoleae n.sp.), four new species of Namanereis (N. minuta n.sp., N. serratis n.sp., N. stocki n.sp., N. sublittoralis n.sp.), and three widespread species groups ( Namalycastis abiuma, Namanereis littoralis, N. quadraticeps). Fourteen species are newly placed into synonymy, Lycastis maxillo-falciformis Harms, L. maxillo-ovata Harms, L. maxillo-robusta Harms, Lycastis meraukensis Horst, L. nipae Pflugfelder, L. ouanaryensis Gravier, L. ranauensis Feuerborn, L. vivax Pflugfelder, Lycastopsis augeneri Okuda, L. tecolutlensis Rioja, Namalycastis rigida Pillai, N. tachinensis Rosenfeldt, N. vuwaensis Ryan, and Namanereis littoralis Hutchings & Turvey. A neotype is designated for Namalycastis hawaiiensis (Johnson), and lectotypes are designated for Namalycastis geayi (Gravier), N. senegalensis (Saint-Joseph), N. terrestris (Pflugfelder), Namanereis amboinensis (Pflugfelder) and N. littoralis (Grube). Keys to genera and species are given. Namanereidinae are generally confined to the tropics and subtropics. Maximum species-diversity occurs in the Caribbean and Indo-Pacific, in particular in coastal areas subjected to recent uplifting, where both littoral-zone and freshwater (riparian and subterranean) forms occur. Phylogenetic results indicate that in both Namalycastis and Namanereis there is a preference for freshwater habitats among species with apomorphic traits (corollary being that marine habitats are favoured by the plesiomorphic members). This suggests that the ancestor of the Namanereidinae was a euryhaline coastal species.

A Dynamic Adaptive Relaxation Scheme Applied to the Euclidean Steiner Minimal Tree Problem

The Steiner problem is an NP-hard optimization problem which consists of finding the minimal-length tree connecting a set of N points in the Euclidean plane. Exact methods of resolution currently available are exponential in N, making exact minimal trees accessible for only small size problems (up to N \approx 100). An acceptable suboptimal solution is provided by the minimum spanning tree (MST) which has been shown computable in an O(N log N) step. We propose here an O(N) process that is able to relax a given initial Steiner tree into a local minimum of its length. This process, based on a physical analogy, simulates the dynamics of a fluid film which relaxes under surface tension forces and stabilizes in an equilibrium configuration minimizing its total length, through purely local interactions. To improve the solution to the Steiner problem, this O(N) relaxation scheme is applied to reduce the length of the MST. This results in a heuristic of a very low O(N log N) complexity for the Steiner problem, whose performance is shown to compare quite favorably with that of the best available heuristics. Large problem sizes up to N=10000 were successfully tackled. A characterization of the asymptotic behavior of the solution of the Steiner problem shows a stabilization to a nonvanishing positive value of the average length reduction achieved over the MST and predicts an average length for the minimal Steiner tree of about 3% below 0.65 N1/2 for large N.

Parallel algorithms for the minimum cut and the minimum length tree layout problems

The minimum cut and minimum length linear arrangement problems usually occur in solving wiring problems and have a lot in common with job sequencing questions. Both problems are NP-complete for general graphs and in P for trees. We present here two parallel algorithms for the CREW PRAM. The first solves the minimum length linear arrangement problem for trees and the second solves the minimum cut arrangement for trees. We prove that the first problem belongs to NC for trees, and the second problem is in NC for bounded degree trees. To the best of our knowledge, these are the first parallel algorithms for the minimum length and the minimum cut linear arrangement problems.

The Calliphoridae (Blowflies) (Diptera: Oestroidea) are Not a Monophyletic Group

To investigate the monophyly and phylogenetic relationships of the Calliphoridae (blowflies), 23 terminal monophyletic taxa within the Oestroidea and two outgroup taxa (Muscidae and Anthomyiidae) were scored for 45 adult and larval characters and subjected to analyses using parsimony. Search for trees of maximum total fit (using Pee-Wee) gave nine different trees for the possible values of the concavity constant, whereas search for minimum length trees under equal weights (using NONA and PAUP) gave two, different, trees. The preferred trees are those obtained from the maximum-fit searches since they are derived from a method that lowers the weight of characters showing within-terminal as well as within-tree homoplasy. The family Calliphoridae in various commonly used senses (excluding Oestridae s.lat.; including the Rhiniinae; excluding or including the Rhinoporidae; excluding or includingMystacinobia) does not appear to be a monophyletic group. It appears as such only in trees having a much lower total fit than the optimal trees (maximum fit approach) or only in a very small fraction of the numerous trees five or nine steps longer than the minimum length trees. The assemblage (Rhiniinae+Toxotarsinae+Chrysomyinae), sometimes assumed to be a natural taxon on account of common possession of a row of setae on the upper surface of the stem-vein, should likewise be rejected. A clade (Ameniini+Euphumosia+Phumosiinae+Mesembrinellinae+Parameniini+Catapicephala) is possibly Gondwanian. A clade (Toxotarsinae+Chrysomyinae+Calliphorinae+Luciliinae+Melanomyinae) contains all the sarcosaprophagous “blowflies”. The group (Tachinidae+Sarcophagidae) is possibly the monophyletic sister group of the Rhiniinae.Mystacinobiaemerges from the analyses as the sister group of all other Oestroidea, and the family Axiniidae as the sister group of all the other non-mystacinobiine oestroids. AlthoughMystacinobiais a guano-feeder in all stages, the biology of the other basal oestroid taxa indicates that a reproductive association with living or dead invertebrates is the primitive oestroid pattern. Fossils having a bearing upon the estimation of the age of various oestroid clades are discussed.

The Calliphoridae (Blowflies) (Diptera: Oestroidea) are Not a Monophyletic Group1.

To investigate the monophyly and phylogenetic relationships of the Calliphoridae (blowflies), 23 terminal monophyletic taxa within the Oestroidea and two outgroup taxa (Muscidae and Anthomyiidae) were scored for 45 adult and larval characters and subjected to analyses using parsimony. Search for trees of maximum total fit (using Pee-Wee) gave nine different trees for the possible values of the concavity constant, whereas search for minimum length trees under equal weights (using NONA and PAUP) gave two, different, trees. The preferred trees are those obtained from the maximum-fit searches since they are derived from a method that lowers the weight of characters showing within-terminal as well as within-tree homoplasy. The family Calliphoridae in various commonly used senses (excluding Oestridae s.lat.; including the Rhiniinae; excluding or including the Rhinoporidae; excluding or includingMystacinobia) does not appear to be a monophyletic group. It appears as such only in trees having a much lower total fit than the optimal trees (maximum fit approach) or only in a very small fraction of the numerous trees five or nine steps longer than the minimum length trees. The assemblage (Rhiniinae+Toxotarsinae+Chrysomyinae), sometimes assumed to be a natural taxon on account of common possession of a row of setae on the upper surface of the stem-vein, should likewise be rejected. A clade (Ameniini+Euphumosia+Phumosiinae+Mesembrinellinae+Parameniini+Catapicephala) is possibly Gondwanian. A clade (Toxotarsinae+Chrysomyinae+Calliphorinae+Luciliinae+Melanomyinae) contains all the sarcosaprophagous "blowflies". The group (Tachinidae+Sarcophagidae) is possibly the monophyletic sister group of the Rhiniinae.Mystacinobiaemerges from the analyses as the sister group of all other Oestroidea, and the family Axiniidae as the sister group of all the other non-mystacinobiine oestroids. AlthoughMystacinobiais a guano-feeder in all stages, the biology of the other basal oestroid taxa indicates that a reproductive association with living or dead invertebrates is the primitive oestroid pattern. Fossils having a bearing upon the estimation of the age of various oestroid clades are discussed.

A central limit theorem for the parsimony length of trees

In phylogenetic analysis it is useful to study the distribution of the parsimony length of a tree under the null model, by which the leaves are independently assigned letters according to prescribed probabilities. Except in one special case, this distribution is difficult to describe exactly. Here we analyze this distribution by providing a recursive and readily computable description, establishing large deviation bounds for the parsimony length of a fixed tree on a single site and for the minimum length (maximum parsimony) tree over several sites. We also show that, under very general conditions, the former distribution converges asymptotically to the normal, thereby settling a recent conjecture. Furthermore, we show how the mean and variance of this distribution can be efficiently calculated. The proof of normality requires a number of new and recent results, as the parsimony length is not directly expressible as a sum of independent random variables, and so normality does not follow immediately from a standard central limit theorem.

A central limit theorem for the parsimony length of trees

In phylogenetic analysis it is useful to study the distribution of the parsimony length of a tree under the null model, by which the leaves are independently assigned letters according to prescribed probabilities. Except in one special case, this distribution is difficult to describe exactly. Here we analyze this distribution by providing a recursive and readily computable description, establishing large deviation bounds for the parsimony length of a fixed tree on a single site and for the minimum length (maximum parsimony) tree over several sites. We also show that, under very general conditions, the former distribution converges asymptotically to the normal, thereby settling a recent conjecture. Furthermore, we show how the mean and variance of this distribution can be efficiently calculated. The proof of normality requires a number of new and recent results, as the parsimony length is not directly expressible as a sum of independent random variables, and so normality does not follow immediately from a standard central limit theorem.

The preoral cavity of lower Hymenoptera (Insecta): comparative morphology and phylogenetic significance

The skeletal and muscular morphology of the preoral cavity, including the labrum, hypopharynx and labium, was examined in the imago in representatives of all the ‘symphytan’ families as well as the apocritan families Stephanidae, Megalyridae and Trigonalyidae.Xyelidae have complex modifications for masticating pollen, remarkably similiar to those of primitive Lepidoptera. These modifications, collectively termed the triturating basket complex, include an asymmetrical distal epipharyngeal wall with a microtrichial brush and an enlarged infrabuccal pouch with heavy cuticular armature that interacts with the mandibles during feeding. There were striking structural differences between the two subfamilies of Xyelidae in the ligular region; the reduced glossa and clubshaped paraglossae of Macroxyelinae resembles those of primitive Lepidoptera, while the well developed, flattened glossa and paraglossae in Xyelidae are similiar to those of most other ‘Symphyta’.A putative transformation series, leading from a relatively large labrum with unsclerotised distal epipharyngeal wall lying anterior to the mandibles, as seen in Xyelidae and enthredinoidea, to a small and heavily sclerotised labrum and distal epipharyngeal wall lying posterior to the mandibles, as seen in ‘Siricoidea’, Orussidae and the Apocrita, was revealed. These modifications may be adaptations to enable the adult of the families pupating in wood to emerge from the pupal chamber.The Anaxyelidae, Orussidae and Apocrita have similiar configurations of the glossa and nsertions of the ventral premental adductors. This indicates a close affinity of the Anaxyelidae to Orussidae + Apocrita, a hypothesis that is in conflict with other character systems. The Orussidae and Stephanidae share a unique condition in the development of a pair of large apodemes attached to the labrum; this renders the groundplan state of the labrum in the Apocrita uncertain.Twentyfive characters were defined in an attempt to eludicate the ‘Symphyta’–Apocrita transition. A numerical cladistic analysis of the characters was undertaken, resulting in 522 minimum length trees. The characters are also discussed with reference to a cladogram which resulted from an analysis of the characters derived from the present study and a survey of characters from literature.

Phylogeny of Polemoniaceae Based on Nuclear Ribosomal Internal Transcribed Spacer DNA Sequences

Nuclear ribosomal internal transcribed spacer (ITS) DNA sequences are used to estimate the phylogeny of 53 members of Polemoniaceae, representing all but two genera of the family. Fitch parsimony analysis of equal-weighted nucleotide sites result in 1080 minimal-length trees. However, when alignment-ambiguous positions are removed and an II: 10 transition to transversion weighting is imposed only eight trees are found. These data are used to address two issues: I) patterns of diversification in Polemoniaceae, and 2) the circumscription and monophyly of the genus Gilia. Although the monophyly of Polemoniaceae is well supported, relationships inferred among the earliest diverging lineages are altered by character weighting, treatment of indels, and taxon inclusion. In spite of the lack of reliable resolution at the basal nodes, ITS data provide evidence that Gilia, as currently interpreted, is polyphyletic and comprises at least five independent lineages.

The articulation of annelids

The aim of this paper is to assess the monophyly of the Annelida. Also, recent cladistic analyses of metazoan taxa, using a variety of data, have shown incongruities with regards to annelids and associated taxa that should be resolved. The Platyhelminthes is selected as the taxon to root our minimal length trees and polarise our characters in a parsimony analysis; ingroup taxa being Mollusca, Nemertea, Sipuncula, Echiura, Pogonophora, Vestimentifera, Euarthropoda, Onycho‐phora, and the groups most commonly regarded as true ‘annelids’, the Clitellata and Polychaeta. We use 13 characters and a total of 33 states. This results in 18 minimal length trees of 23 steps. The consensus tree has the topology (Platyhelminthes (Nemertea (Sipuncula Mollusca (Echiura (Polychaeta (Vestimcntifera Pogonophora) Clitellata (Euarthropoda Onychophora)))))). The name Articulata is applied to the Clitellata, Euarthropoda, Onychophora, Pogonophora, Polychaeta, and Vestimentifera. The Vestimentifera is the sister group to, or more likely a clade within, the frenulate pogonophores, and the name Pogonophora is retained for this group. In half of the 18 minimal length trees, the traditionally formulated Annelida, i.e. Polychaeta and Clitellata, is paraphyletic if the Pogonophora are excluded. In the remaining minimal length trees, a monophyletic Annelida cannot be formulated. The name Annelida should not be used unless relationships within the Articulata are resolved to show it is a monophyletic taxon. The taxon name Articulata, originally formulated to include the Annelida and Arthropoda by Cuvier, is defined as the clade stemming from the first ancestor to show repetition of homologous body structures derived by teloblastic growth with a pygidial growth zone (segmentation) and longitudinal muscles broken into bands. The Articulata is considered, on current evidence, to consist of four monophyletic groups; the Arthropoda, Clitellata, Polychaeta, and Pogonophora, though the latter group may be a clade of polychaetes. If this is shown, the Pogonophora should revert to the original family name Lamellisabellidae Uschakov, 1933. An indented classification reflective of the cladistic pattern is provided. Other recent hypotheses about metazoan systematics arc analysed.

Phylogeny, Molecules Versus Morphology, and Rates of Character Evolution Among Fruitbats (Chiroptera: Megachiroptera)

Andersen's 1912 monograph on megachiropterans remains the definitive work on the systematics of this group. Andersen argued that the Macroglossinae, containing the eonycterine and notopterine sections, are a monophyletic sister-group to other fruitbats (i.e. Andersen's Rousettus, Cynopterus and Epomophorus sections). Two recent molecular studies (DNA hybridisation and restriction mapping of ribosomal cistrons), as well as an analysis of female reproductive characters, challenge the monophyly of the Macroglossinae and several of Andersen's other conclusions such as the phylogenetic position of Nyctimene. We performed a cladistic analysis on 36 morphological characters, including 33 that were gleaned from Andersen, to determine whether phylogenetic hypotheses based on modem phylogenetic methods are in agreement with Andersen's original conclusions and to compare morphological and molecular phylogenetic hypotheses. Minimum-length trees based on parsimony are largely consistent with Andersen and support (1) a monophyletic Macroglossinae, within which the eonycterine section is paraphyletic with respect to a monophyletic notopterine section, (2) a monophyletic Cynopterus section, excepting the exclusion of Myonycteris, (3) a monophyletic Epomophorus section, excepting the exclusion of Plerotes, and (4) a paraphyletic Rousettus section, with several of the Rousettus-like forms branching off near the base of the tree. Bootstrapping analyses on a reduced data-set that included taxa shared in common with the DNA hybridisation study did not provide strong support (greater than or equal to 95%) for any clades but did provide moderate support (greater than or equal to 70) for several clades, including a monophyletic Macroglossinae. These findings are in marked contrast to the DNA hybridisation phylogeny. A high index of between-data-set incongruence is further evidence for the clash between DNA hybridisation and morphology. A phylogenetic framework was constructed on the basis of morphological data and DNA hybridisation data using a criterion of moderate support and shows little resolution, whereas employing a criterion of strong support produced a framework resolving several additional nodes. One implication of this framework is that characteristic macroglossine features such as a long tongue with a thick carpet of filiform papillae have evolved independently on several occasions (or evolved once and were lost several times). Rates of character evolution for the morphological characters employed in our analysis were calculated using divergence times estimated from DNA hybridisation data. Rates have apparently been fastest in the interior branches, and slower along the external branches, which suggests an early adaptive radiation in the history of fruitbats.

A molecular phylogeny of Arceuthobium (Viscaceae) based on nuclear ribosomal DNA internal transcribed spacer sequences

Ribosomal DNA (rDNA) internal transcribed spacer (ITS) and 5.8S rDNA sequences were obtained from 22 species of dwarf mistletoes (Arceuthobium — Viscaceae) to test phylogenetic relationships. Interspecific distances ranged from 0 to 21.4% between New World species, values two to five times higher than those measures for the ITS region in other plants. One Old World species (A. oxycedri) and one New World species (A. abietis‐religiosae) were remarkably similar to each other but exhibited up to 41% sequence divergence from the remaining species. Minimum length trees support the concept of a verticillately branched subgenus Arceuthobium; however, interspecific distances indicate this group is extremely heterogeneous. Subgenus Vaginata, Section Vaginata, is centered in Mexico and encompasses all the taxa previously placed in this group but is expanded to include several species previously classified in Section Campylopoda (e.g., A. divaricatum, A. rubrum, and A. strictum). The sister group relationship between A. divaricatum and A. douglasii, first seen following isozyme analysis, is supported by ITS sequence data. Section Campylopoda s. s. is now composed of 13 mainly U.S. species that show a high degree of morphological and genetic similarity. The eastern dwarf mistletoe, A. pusillum, is not closely related to A. douglasii but rather with A. bicarinatum from Hispaniola, which suggests that these taxa represent highly modified relicts that shared an ancestor in the early Tertiary. Two endemic species from Mexico and Central America (A. guatemalense and A. pendens) formed a sister group and have been placed in a new Section (Penda). Rapid molecular evolution in Arceuthobium may be associated with the adaptive radiation of this genus on numerous conifer hosts.

A Molecular Phylogeny of Arceuthobium (Viscaceae) Based on Nuclear Ribosomal DNA Internal Transcribed Spacer Sequences

Ribosomal DNA (rDNA) internal transcribed spacer (ITS) and 5.8S rDNA sequences were obtained from 22 species of dwarf mistletoes (Arceuthobium — Viscaceae) to test phylogenetic relationships. Interspecific distances ranged from 0 to 21.4% between New World species, values two to five times higher than those measures for the ITS region in other plants. One Old World species (A. oxycedri) and one New World species (A. abietis-religiosae) were remarkably similar to each other but exhibited up to 41% sequence divergence from the remaining species. Minimum length trees support the concept of a verticillately branched subgenus Arceuthobium; however, interspecific distances indicate this group is extremely heterogeneous. Subgenus Vaginata, Section Vaginata, is centered in Mexico and encompasses all the taxa previously placed in this group but is expanded to include several species previously classified in Section Campylopoda (e.g., A. divaricatum, A. rubrum, and A. strictum). The sister group relationship between A. divaricatum and A. douglasii, first seen following isozyme analysis, is supported by ITS sequence data. Section Campylopoda s. s. is now composed of 13 mainly U.S. species that show a high degree of morphological and genetic similarity. The eastern dwarf mistletoe, A. pusillum, is not closely related to A. douglasii but rather with A. bicarinatum from Hispaniola, which suggests that these taxa represent highly modified relicts that shared an ancestor in the early Tertiary. Two endemic species from Mexico and Central America (A. guatemalense and A. pendens) formed a sister group and have been placed in a new Section (Penda). Rapid molecular evolution in Arceuthobium may be associated with the adaptive radiation of this genus on numerous conifer hosts.

228 DIVERSITY AND PHYLOGENETIC RELATIONSHIPS OF JUGLANS SPECIES DETERMINED FROM NUCLEAR DNA RFLP ANALYSIS

RFLPs were studied in 41 populations of 13 Juglans species to study genetic diversity and phylogenetic relationships. 19 single locus nuclear RFLP loci were used to generate genetic distance/identity matrices based on allele frequencies. 21 probes were used to generate genetic distances and phylograms using shared-fragments with parsimony analysis. Parsimony analysis on fragment data produced a minimal length tree in general agreement with distance data trees, but with additional phylogenetic resolution resembling previous systematic studies. All analyses indicate an ancient origin of J. regia, which has been considered a recently derived species. A 10x difference in heterozygosity was seen among species. Genetic differentiation among conspecific east Asian populations was larger than among east Asian species. The opposite was true for American species. J. hindsii is classified as a distinct species and J. cinerea was included in section Cardiocaryon rather than Trachycaryon, from the diversity analysis.