HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

What's this?

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (43) Citing Articles via Google Scholar Google Scholar Articles by Goffinet, B. Articles by Vitt, D. H. Search for Related Content PubMed Articles by Goffinet, B. Articles by Vitt, D. H. Agricola Articles by Goffinet, B. Articles by Vitt, D. H. Social Bookmarking                            What's this?

Circumscription and phylogeny of the Orthotrichales (Bryopsida) inferred from RBCL sequence analyses¹

^a Department of Biological Sciences, The University of Alberta, Edmonton, Alberta, Canada T6G 2E9

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The affinities as well as the circumscription of the Orthotrichaceae (Bryopsida), one of the most diverse families of mosses, have been the focus of a controversy for much of the last century. We obtained rbcL sequences for 37 arthrodontous mosses, including 27 taxa of the Orthotrichales. The sequences were analyzed using maximum parsimony and neighbor joining in order to (1) test the monophyly of the Orthotrichales and the Orthotrichaceae; (2) determine their phylogenetic relationships; and (3) test the current subfamilial classification within the Orthotrichaceae. Both analyses suggest that the Orthotrichales are polyphyletic. The Erpodiaceae and the Rhachitheciaceae as well as Amphidium and Drummondia, two genera of the Orthotrichaceae, are shown to be of haplolepideous affinity. The Splachnales, the Bryales sensu lato, and the Orthotrichales form a monophyletic clade sister to the Haplolepideae. Both neighbor joining and maximum parsimony also suggest that the Orthotrichaceae are composed of two major lineages dominated either by acrocarpous or cladocarpous taxa. The monophyly of the family is, however, only well supported by Tamura's distances. The genera Macrocoma, Macromitrium, Orthotrichum, Ulota, and Zygodon all appear to be artificial assemblages. This study illustrates the contribution of rbcL sequence data to bryophyte systematics and, particularly, in determining the affinities of taxa lacking a peristome, whose characters are central to the classification of mosses.

Key Words: bryophytes • evolution • mosses • Orthotrichales • Orthotrichaceae • phylogeny • rbcL • systematics.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The central concept in the Hennigian phylogenetic approach is the use of appropriate outgroups, which allows adequate polarization of character state transformation in the ingroup. A reconstruction of the evolutionary history of the Orthotrichaceae, one of the most diverse families of mosses, based on morphology has been hampered by the lack of a peristome in certain taxa, the absence of gametophytic characters that are informative at higher taxonomic levels, and consequently the absence of a consensus regarding the putative sister group and other outgroups to the order. Within the Bryophyta sensu lato, phylogenetic reconstructions using nucleotide sequence data have focused exclusively on the monophyly and the relationships of the division and its classes (e.g., Mishler et al., 1992, 1994; Waters et al., 1992; Capesius, 1995; Bopp and Capesius, 1996; Hedderson, Chapman, and Rootes, 1996) rather than on the evolutionary history of arthrodontous mosses.

In arthrodontous mosses (Bryopsida; sensu Vitt, Goffinet, and Hedderson, 1997), the peristome teeth that line the mouth of the capsule and regulate spore dispersal are made of cell wall remnants. Three concentric layers, namely the outer (OPL), the primary (PPL), and the inner (IPL) peristomial layer, contribute to the teeth (Blomquist and Robertson, 1941). The number and the pattern of cell divisions occurring in these layers, particularly in the IPL, are central to the classification of arthrodontous mosses. Vitt (1984) recognized five types of articulate peristomes (i.e., four diplolepideous and one haplolepideous type), which characterize the subclasses of the Bryopsida (see also Edwards, 1979, 1984). Vitt, Goffinet, and Hedderson (1997) recently argued that morphological characters used to define these major peristome types are not phylogenetically informative and that higher level phylogenetic reconstruction of the Bryopsida will rely on independent sources of data such as ontogeny and gene sequences. The development of the funariaceous, bryaceous, and dicranaceous peristome types has been studied recently (Shaw, Anderson, and Mishler, 1989; Shaw, Mishler, and Anderson, 1989; Schwartz, 1994), but unless similar data are available for the orthotrichaceous peristome, polarizing the developmental pathways and thus defining homologies and monophyletic groups based on peristome architecture remain impossible.

The peristome of the Orthotrichales is diplolepideous and is characterized by the following features: «an endostome with segments that alternate with exostome teeth; lack of basal membrane and with segments, which are not or are rarely keeled; and an exostome that has a thick, outer layer and a thin, inner layer» (Vitt, 1982a). Crosby (1980) and Vitt (1981a) were the first to suggest that the ancestral arthrodontous peristome was diplolepideous. Crosby (1980) and later Shaw and Rohrer (1984) considered that the Bryum type represented the most primitive among the diplolepideous types and consequently that the orthotrichaceous peristome was derived through reduction, a hypothesis also supported by Hedenäs (1994). Vitt (1984) argued that «the evolution of the peristome has not been uni-directional» and that the bryaceous peristome «is a totally separate evolutionary line than either the orthotrichaceous or haplolepideous divergences» from an ancestral funariaceous type (see also Vitt, 1981a). More recently, Lewinsky (1989) argued in favor of a distinct orthotrichaceous peristome type that is «most likely to have evolved from a peristome with a formula of 4:2:4 with complete alignment of the cells in the IPL,» thus the funariaceous type, but unlike Vitt (1984) she considers the orthotrichaceous peristome as possibly pivotal to the evolution of the dicranaceous and the bryaceous types.

The Orthotrichaceae sensu Vitt (1984; and not sensu Churchill and Linares, 1995) include over 500 species, distributed among 22 genera (see Goffinet, 1997a, b), with Orthotrichum, and Macromitrium accounting for more than two-thirds of the species richness (Vitt, 1982a). The family is cosmopolitan in distribution and is particularly prevalent in tropical montane forests. The Orthotrichaceae are traditionally characterized by a unique peristomial architecture (see above), as well as «small, papillose upper leaf cells; no differentiated alar cells; large, usually mitriform calyptrae; terminal perichaetia with additional growth by lateral innovations» (Vitt, 1982b). These characters are, however, by no means constant within the family, resulting in alternative hypotheses of familial circumscription. The controversy has focused particularly on the haplolepideous vs. the diplolepideous affinity of Amphidium, which lacks a peristome, and has atypical cell ornamentation (Brotherus, 1909 vs. 1925; Vitt, 1973 vs. Lewinsky, 1976), or of taxa whose peristome is reduced (Drummondia: Shaw, 1985 vs. Vitt, 1972). The current infrafamilial classification (Vitt, 1982a) reflects the four possible combinations of two characters, position of the female gametangia and «shape» of the calyptrae, and deviates little from Brotherus's concept early this century (Brotherus, 1925). In Vitt's (1982a) phylogenetic arrangement, the Zygodontoideae (Schimp.) Broth. (acrocarpous and cucullate calyptrae) are basal, followed by the Drummondioideae Vitt (cladocarpous and cucullate calyptrae), which are sister to a clade composed of the Orthotrichoideae Broth. and the Macromitrioideae Broth. (acrocarpous and cladocarpous, respectively, and both with mitrate calyptrae).

Addressing the phylogenetic relationships of the Orthotrichales and its delineation is essential for reconstructing the evolutionary history of orthotrichaceous genera. Goffinet (1997b) recently excluded the Microtheciellaceae from the Orthotrichales based on their pleurocarpy. Goffinet (1997a) also transfered two genera from the Orthotrichaceae to the Rhachitheciaceae. The peristomial architecture of the Rhachitheciaceae (Goffinet, 1997a) and of the Erpodiaceae Broth. (Crum, 1972; Edwards, 1979) suggests possible affinities to the Haplolepideae, and particularly the Seligeriales. The monophyly of the Orthotrichaceae and the Orthotrichales has recently also been questioned by De Luna (1995) while examining the systematic affinities of the Hedwigiaceae. Based on a phylogenetic analysis using morphological characters, De Luna (1995) concluded that the Orthotrichales merely represent an evolutionary grade, suggesting that a clade of predominantly cladocarpous Orthotrichaceae (Macromitrioideae, Drummondioideae) and the putatively related Erpodiaceae Broth., Microtheciellaceae Harrington and Miller, and Helicophyllaceae Broth.(see below) was sister to the pleurocarpous mosses, and separated from the acrocarpous Orthotrichaceae and their possible sister group, the Rhachitheciaceae Robins. by the Hedwigiaceae. Affinities between the Hedwigiaceae and the Orthotrichaceae have been proposed by other authors (Walther, 1983; Frey et al., 1995). Because of the gymnostomous nature of the Hedwigiaceae and a unique combination of gametophytic characters (acrocarpy, lack of costa, and presence of pseudoparaphyllia), the affinities of Hedwigia are unlikely to be resolved with traditional morphological characters.

Severe reduction trends in morphological characters, particularly with regard to the peristome, are thus the source of the controversies regarding the delineation of the Orthotrichales and its type family, the Orthotrichaceae. Ambiguous homologies between characters defining alternative peristome types hamper resolving the phylogenetic affinities of these taxa. Variation in the nucleotide sequence of the chloroplast gene rbcL, encoding for the ribulose 1,5 biphosphate carboxylase, has found a wide application in reconstructing the evolutionary histories within suprageneric taxa of vascular plants (Chase et al., 1993; Hasebe et al., 1995). The present study aims at setting the foundations for reconstructing the phylogeny of all orthotrichaceous genera based on morphology and for addressing the character evolution in this taxonomically and morphologically diverse family. Thus, the objectives were to use rbcL sequence data (1) to test the monophyly of the Orthotrichales and the Orthotrichaceae; (2) to determine the phylogenetic position of the Orthotrichales within the arthrodonteae, and thus identify its putative sister group; and (3) to test the monophyly and the phylogenetic relationships of the subfamilies proposed by Vitt (1982a).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Taxon sampling
Following the exclusion of five genera from the Orthotrichaceae (Goffinet, 1997a, b) the family now consists of 22 genera. While material has been seen for all taxa except for the recently described Orthomitrium (Lewinsky-Haapasaari and Crosby, 1996), extractions were not attempted for taxa known only from the type specimen (Ceuthotheca Lewinsky-Haapasaari, Leiomitrium Mitt., Leratia Broth.), or for which only very little material was available (Stoneobryum Norris and Robinson). DNA extractions were attempted for the remaining 17 genera, and for the most diverse genera, such as Macromitrium and Orthotrichum, several species belonging to morphologically distinct infrageneric taxa were tentatively included. DNA extractions were also attempted for representatives of all related families in the Orthotrichales. Furthermore, ten outgroup taxa were chosen, representing the Funariaceae, Splachnaceae, Encalyptaceae, Hedwigiaceae, the Haplolepideae (Ptychomitriaceae and Rhabdoweisiaceae), and the ciliate alternate peristomate mosses (Mniaceae and Thuidiaceae).

DNA extraction and PCR-DNA amplification
DNA was extracted, following a modification of Doyle and Doyle (1987), from apices of stems and branches, or single sporophytes sampled from dried herbarium or fresh collections. Gametophytic samples were cleaned of contaminating material in distilled water, then air-dried overnight and lyophilized prior to the extraction. One to 10 mg dry mass of each sample were placed in a spot plate, covered with liquid N₂, and ground to a fine powder once the liquid N₂ was completely evaporated. The powder was then suspended in 250 µL of 2X CTAB (hexadecyltrimethylammonium bromide)-1% beta-mercapto-ethanol extraction buffer heated at 60°C and the solution transferred to a microcentrifuge tube. The tubes were then placed in a 60°C water bath. After 45 min, 250 µL of CI (chloroform-isoamyl 24:1) were added and the tubes gently inverted for 10 min. Cell debris were then isolated by centrifuging the tubes for 10 min at maximum speed (e.g., 14000 rpm). The aqueous phase was transferred to a fresh tube (the CI extraction can be repeated if the aqueous solution is not clear). Then 350 µL of 95% ethanol were added and the DNA allowed to precipitate at 4°C. After at least 2–3 h, the tubes were centrifuged for 5 min at maximum speed again to pellet the DNA. The ethanol was discarded and the pellet washed gently by adding 350 µL of 70% ethanol. The tubes were centrifuged again for 5 min at 7000 g. The ethanol was discarded and the pellet dried in a vacuum centrifuge. The DNA was then suspended in 100 µL of TBE (Tris-boric acid-EDTA) at 65°C for 10 min and kept at -20°C subsequently. The rbcL gene was amplified via the polymerase chain reaction (PCR) using Taq Polymerase (Promega Corporation, Madison, WI). The amplification reaction was performed in 50 µL volume including 50 mmol/L KCl, 10 mmol/L Tris-HCl, 0.1% Triton, 5% glycerol, 2.5 mmol/L MgCl₂, 20 mmol/L of each dNTP, 0.2 mmol/L of primers Z₁ and 1351R (Wolf, Soltis, and Soltis, 1994) and 0.2–30.0 ng of template DNA. The samples were exposed to the following temperature profiles using a Grant thermal cycler: one cycle of 94°C for 3 min and 85°C for 4 min, during which 1 unit of Taq DNA Polymerase would be added to each tube, and 30 cycles of 94°C for 1 min, 52°C for 1 min, 72°C for 2 min, and finally one segment of 72°C for 10 min. The double-stranded product subsequently served as the template for the amplification of single-stranded target DNA. This second PCR followed the same profiles as before, but the reaction solution included only one of the two PCR primers: Z₁ for amplification of the forward strand and 1351R for the reverse strand. The single-stranded product was precipitated with 20% PEG/2.5 mol/L NaCl, washed first with 70% EtOH and then with 95% EtOH, before being suspended in 7 mL of Tris buffer-EDTA (Morgan and Soltis, 1993).

Sequencing and alignment
Sequencing the single-stranded template followed the dideoxy chain termination method (Sanger, Nicklen, and Coulson, 1977) using the Sequenase;rm version 2.0 T7 DNA polymerase following the manufacturer's instructions (Amersham, Canada). The sequence of the primers used is given in Table 1. The sequencing products were electrophoresed on a 6% polyacrylamide gel (0.4 mm thickness; 1X TBE buffer), at 2400 V for 4 h. The gels were fixed in 10% acetic acid, washed with distilled water, and dried in an oven at 65°C for 30 min and autoradiographed for 36–48 h. The sequences were entered in MacClade (version 3.03) and aligned manually against available sequences of Sphagnum palustre and Andreaea rupestris (GenBank accession numbers L13485 and L13473, respectively; Manhart, 1994). The first and last 30 nucleotide sites, corresponding to the sequences of the PCR primers, were excluded from parsimony analyses.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The rbcL gene was successfully sequenced for 22 taxa of the Orthotrichaceae distributed among ten genera, for representatives of two of the related families as well as for all selected outgroup taxa (Table 2). Amplification products were not obtained for Florschuetziella Vitt, Leptodontiopsis Broth., Muelleriella Dusáen, Pleurorthotrichum Broth., and Stenomitrium (Mitt.) Broth., Helicophyllum Brid. The sequences obtained from PCR fragments generated using the PCR primers Z₁ and 1351R were, as expected, 1320 bases long (±90% of the total length of the gene). Alignment of the sequences with known sequences of Sphagnum pallustre and Andreaea rupestris did not require the inclusion of gaps. Phylogenetically informative characters are preponderant at the third codon position followed by the first and the second codon site (Table 3). RbcL sequence variation yields 108 sites that are potentially phylogenetically informative within the Orthotrichaceae. Informative characters are distributed fairly evenly across the sequence (Fig. 1). The distribution of 500000 random trees using all characters had a skewness index g₁ of -0.55; excluding the first and second codon position yielded a similar index (g₁ = -0.54), while trees generated solely from characters of these two first codon sites had a higher skewness index (g₁ = -0.39).

View larger version (67K): [in this window] [in a new window]   Fig. 1. Distribution of phylogenetically informative (PI) characters, the average consistency index (CI), and the average number of steps per phylogenetically informative character along the rbcL sequence based on tree shown in Fig. 3. Each class contains 50 nucleotide sites, except for the first class that includes only sites 31–50. Cross-lined bars = total number of phylogenetic informative characters; solid bars = average CI multiplied by 10; hollow bar = total number of steps per total number of phylogenetically informative characters.

View larger version (45K): [in this window] [in a new window]   Fig. 2. Summary consensus trees of most parsimonious Fitch trees found in a heuristic search using rbcL sequence data with Sphagnum palustre and Andreaea rupestris as outgroups. (A): strict consensus of all three islands (39 MPTs; tree 964 steps; CI: 0.390; RI: 0.624); (B): 50% majority rule tree of island 3 (12 MPTs; mean f value: 13 943 ± 656); (C): 50% majority rule tree of islands 1 and 2 (27 MPTs; mean f value: 16 633 ± 1 308); (D): 50% majority rule tree of all 15 MPTs showing the Orthotrichaceae monophyletic (mean f value: 16722 ± 2015).

View larger version (45K): [in this window] [in a new window]   Fig. 3. One of 39 most parsimonious Fitch trees found based on rbc L sequence data and using Sphagnum palustre and Andreaea rupestris as outgroups. The dotted lines identify branches not present in the strict consensus tree. Bootstrap values (% of 100 replicates) 50% or higher are given below the branch, and decay indexes are presented above the branch. Taxa excluded from the Orthotrichales are in boldface.

View larger version (42K): [in this window] [in a new window]   Fig. 4. Phylogenetic tree reconstructed using the neighbor-joining method, with Tamura's distance parameter including both transitions and transversions. Bootstrap values (% of 100 replicates) 50% or higher are plotted on the tree. Taxa excluded from the Orthotrichales are in boldface. Abbreviations for generic names in the Orthotrichaceae follow those of Table 4.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
RbcL sequence data
RbcL sequence data are routinely used to reconstruct evolutionary histories among (Chase et al., 1993; Hasebe et al., 1995) or more rarely within (e.g., Haufler and Ranker, 1995) genera of vascular plants. With 30% variable sites, including 12% of sites with changes that may be phylogenetically informative, the variation in the nucleotide sequences of the rbcL gene among moss taxa may yield sufficient characters for reconstructing the evolutionary history of the taxa included in this study. The distribution of informative characters appears rather uniform (Fig. 1) and is similar to that observed in vascular plants, i.e., with no obvious hot spot detectable (Olmstead and Sweere, 1994). The distribution of randomly generated trees has a left-hand skewness with a g₁ value (-0.55) significantly smaller than the critical value (-0.10 or -0.08 for 100 or 250 characters, respectively; P = 0.01) furnished by Hillis and Huelsenbeck (1992), suggesting that our rbcL data set is more structured than a random data set of equal size and that it may contain significant phylogenetic signal (Hillis and Huelsenbeck, 1992). Unlike Conti, Fischbach, and Sytsma's (1993) observation that the third codon position introduces «noise,» a lower g₁ value is observed when the matrix is restricted to the third codon position (-0.54 vs. -0.39), suggesting that the changes at the third codon position are more structured. Analyses of a data set restricted to the third codon positions only yield topologies congruent with those obtained with the complete data set, whereas a search excluding changes at the third codon position results in a consensus tree incongruent with monophyletic concepts of either the ciliate mosses, or the Haplolepideae (results not shown). This observation would suggest that the first two codon sites carry more homoplasies than the third position, which may thus be more informative. Alternatively, the taxon sample may be too disparate for variability at the more conserved positions 1 and 2 to be mostly phylogenetically informative. The significance of the difference between the g₁ value based on our data set and that of a random matrix may, as a result, be due to differences in the frequency of character states rather than in the congruence among characters in both data sets (Källersjöo et al., 1992). This second hypothesis most likely applies to our data since our restricted taxon sample represents major lineages of arthrodontous mosses with a long evolutionary history (Frey, 1977, 1990).

An indication of the accuracy of our sequences as well as for the presence of pseudogenes may be obtained from the comparison of amino acid sequences with the distribution of active sites (Kellog and Juliano, 1997). All amino acid residues of the active site found in spinach by crystallography (Andersson et al., 1989) are conserved among studied bryophyte taxa, except for position 404 (amino acid numbering), which is scored as polymorphic (Arg in addition to the conserved Gly) for six taxa, and thus excluded from phylogenetic analyses in MP. This observation may provide some preliminary support for the accuracy of the sequences obtained.

Circumscriptions of the Orthotrichales
The Orthotrichales are currently composed of four families: the Erpodiaceae, Helicophyllaceae, Orthotrichaceae, and Rhachitheciaceae (Goffinet, 1997b). RbcL sequence data analyzed using either the neighbor joining or the maximum parsimony criterion indicate that the Orthotrichales form a polyphyletic taxon. Both the Erpodiaceae (Aulacopilum hodgkinsoniae, Venturiella sinensis) and the Rhachitheciaceae (Uleastrum palmicola) are indeed placed in a clade with haplolepideous taxa (Figs. 3, 4). This clade also includes the «orthotrichaceous» genera Amphidium and Drummondia. The monophyly of this haplolepideous clade is strongly supported (MP: decay index [DI] of 4, bootstrap value [BV] = 77%; NJ: BV = 94%). Constraining these taxa to a relationship within a monophyletic Orthotrichales is furthermore very costly in terms of parsimony, requiring up to 25 additional steps. Variation in the nucleotide sequence of the rbcL gene is thus congruent with earlier hypotheses based on morphology and cytology, suggesting haplolepideous affinities of these taxa (for Amphidium see Anderson and Crum, 1958; Vitt, 1970, 1973; Drummondia: Shaw 1985; Erpodiaceae: Edwards, 1979; Rhachitheciaceae: Goffinet, 1997a). More extensive sampling of haplolepideous taxa would be needed before ordinal affinities of Amphidium and Drummondia are elucidated (see also below, affinities of excluded taxa).

De Luna (1995) recently argued that the Orthotrichales represented an evolutionary grade, reached by two lineages. The cladocarpous lineage (including, e.g., Erpodiaceae, Macromitrioideae) was sister to the Leucodontales and separated from the acrocarpous line (e.g., Rhachitheciaceae, Orthotrichoideae) by the Hedwigiaceae. Affinities between the Hedwigiaceae and the Orthotrichaceae have also been proposed by other authors (Walther, 1983; Frey et al., 1995). Constraining the heuristic search to include Hedwigia ciliata in the Orthotrichaceae (as defined in Figs. 3 and 4, and with no further relationships specified within this clade) yielded 12 trees that not only are 16 steps longer than the MPTs in the unconstrained search, but also share a monophyletic Orthotrichaceae. The incongruence between De Luna's morphological study (De Luna, 1995) and the present molecular analysis may be due to inconsistent and inadequate taxon sampling with the absence of representatives of the Leucodontineae from the molecular analysis and of bryalean taxa from the morphological analysis. The latter analysis further includes taxa that are here shown to be unrelated to the Diplolepideae (i.e., Erpodiaceae, Rhachitheciaceae) and might be affected by the subsequent misinterpretations of the nature of the peristome of these families (scored as an exostome instead of an endostome). A closer relationship between the Hedwigiaceae and the Orthotrichaceae cannot be excluded, but if the Hedwigiaceae were closely related to the Orthotrichaceae, rbcL data suggest that the Orthotrichaceae would most likely remain a natural group. This phylogenetic hypothesis seems to be supported by 18S sequence data analyses (C. Cox and T. Hedderson, personal communication, University of Reading, U.K.). The branch that supports the Hedwigiaceae–cladocarpous Orthotrichales–Leucodontales lineage is supported in De Luna's analysis (1995) by three synapomorphies, namely plagiotropic growth, presence of pseudoparaphyllia, and differentiated perichaetial leaves. Two of these characters are reversed in the cladocarpous Orthotrichales, leaving a single character that actually supports a relationship of these taxa with the Hedwigiaceae and Leucodontales, namely plagiotropic growth. This character is, however, homoplastic among many families and genera of mosses (Meusel, 1935) and may thus not be truly informative at the ordinal level.

Analysis using the distance method yields a single tree (Fig. 4) that shows strong support for the monophyly of the Orthotrichaceae, a phylogeny congruent with that of 15 of the 39 MPTs found in the cladistic analysis. Käss and Wink (1995) and Barker, Linder, and Harley (1995) also used both methods (MP and NJ) for phylogenetic reconstructions based on rbcL sequence data, and the results of both analyses were fairly congruent. Kim, Rohlf, and Sokal (1993) critically examined the accuracy of the NJ method under different constraints on a random data set and found that this method «has the highest accuracy overall.» Russo, Takezaki, and Nei (1996) recently compared the efficiency of various tree-building methods in recovering a known phylogeny and concluded that NJ «gives as good a result as the more time consuming.;th.;th.methods.» The congruence between the NJ tree and 15 MPTs is therefore here interpreted as supporting the monophyly of the Orthotrichaceae.

With the exclusion of the Erpodiaceae, Rhachitheciaceae, and Hedwigiaceae from the Orthotrichales, the order is now reduced to the Orthotrichaceae and the Helicophyllaceae. In the absence of rbcL data, the phylogenetic relationship between these two families remains dubious due to unique gametophytic characters, such as reduced ventral and dorsal leaves, and lateral leaves that are inrolled when dry, and the absence of a peristome in Helicophyllum torquatum, the sole species of the Helicophyllaceae (Vitt, 1982c). Buck and Vitt (1986) argued in favor of a close relationship with the Racopilaceae (ciliate diplolepideae), as suggested by earlier authors (Brotherus, 1909, 1925; Fleischer, 1920; Dixon, 1932; Reimers, 1954). The perichaetia are, however, produced terminally on the main axis (De Luna, 1995), a condition that is a priori incompatible with a placement in a pleurocarpous lineage (La Farge-England, 1996). Other loci are currently being tested for addressing the affinities of the Helicophyllaceae. A putative phylogenetic relationship between the Orthotrichaceae and the Cryphaeaceae and the monotypic Wardiaceae (Walther, 1983) was not examined in the present study. Wardia hygrometrica should be excluded from the Orthotrichales based on well differentiated alar cells and prosenchymateous cells. Combined with the acrocarpous condition (Welch, 1943), these characters may indicate a bryalean origin, a hypothesis recently examined using 18S gene sequences (Hedderson, Cox, and Gibbings, 1997). The Cryphaeaceae have traditionally been placed among the Leucodontineae (Vitt, 1984; Buck and Vitt, 1986). A relationship of this family to the Hedwigiaceae, and thus the Orthotrichales, has been considered and rejected (De Luna, 1995), but may need to be reexamined in the light of the discovery of cladocarpous species of Cryphaea sensu lato (La Farge-England, 1996).

Ordinal relationship of the Orthotrichales
The affinities of the Orthotrichales, here restricted to the Orthotrichaceae, as indicated by comparisons of the rbcL sequences, are within a diplolepideous lineage including the Splachnales and the ciliate mosses. This clade shares a common ancestor with the Haplolepideae, and together they form a sister group to the Funariales (Figs. 3–4). This topology is congruent with Vitt's hypothesis (Vitt, 1981a) that the opposite diplolepideous peristome is primitive among arthrodontous mosses and that the haplolepideous peristome is derived from such an ancestral type (Vitt, 1984). Alternative hypotheses proposed by Lewinsky (1989; Orthotrichales basal to dichotomy between haplolepideae and ciliate diplolepideae) or by Shaw and Rohrer (1984) and Crosby (1980; ciliate diplolepideae are the most primitive mosses) both require five additional steps.

The phylogeny obtained by either NJ or MP analysis (Figs. 3, 4) also agrees with Vitt (1984) with regard to the monophyly of a lineage composed of the Splachnales, Orthotrichales and the ciliate mosses. The relationships among these lineages remain however, ambiguous (Fig. 2). The most parsimonious scenario points toward the Splachnales and the Orthotrichales being sister groups. Vitt (1984), by contrast, proposed that the Orthotrichales are sister to the ciliate mosses, a topology that would only require one additional step based on our rbcL data set. Koponen (1977, 1983) considered the genus Brachymitrion to be the most primitive extant member of the Splachnaceae. Unlike in related genera, the thickening on the PPL is heavier than on the OPL, and the PPL also has strong trabeculae (Koponen, 1977, 1982). Both these features are shared with the Funaria and the Bryum peristome type (Shaw and Rohrer, 1984). If the Orthotrichales and the Splachnales formed a monophyletic group, both orders would most likely remain natural orders, with the Orthotrichales defined by thick-walled laminal cells. The heavier thickening on the OPL could have arisen independently in both lineages or be a synapomorphy for both, with subsequent reversal in Brachymitrion. In terms of peristome evolution, these lineages would form a plesiomorphic sister group to the ciliate mosses, the latter one being defined by the asymmetric division of the IPL leading to the development of cilia. If we consider the alternative scenario where the Splachnales are sister to a clade composed of the Orthotrichales and the ciliate mosses, the most parsimonious topology based on our data suggests that within the latter, the ciliate mosses remain monophyletic. An extensive taxon sampling in the Bryales is, however, needed before their evolutionary relationship can be addressed more critically. The Orthotrichales and the Bryales sensu lato each share one plesiomorphic state with Brachymitrion (Splachnales): either completely aligned cell divisions (Orthotrichales; Lewinsky, 1989) or heavier thickening of the PPL and ventral exostomial trabeculae (typical ciliate mosses). The ciliate mosses have traditionally been defined by a strongly asymmetric division in IPL leading to the development of cilia. While some «ciliate» mosses actually lack cilia (Buck and Vitt, 1986), have a thickened OPL (Eucamptodontopsis, A. Newton, personal communication, Smithsonian), and others even have opposite peristomes (Garovaglia div. sp.; During, 1977; Nishimura and Watanabe, 1992), none has yet been found to have a first-late symmetric division in the IPL. An asymmetric division is elsewhere found only among haplolepideous taxa. Considering that the Rhachitheciaceae are here shown to be of haplolepideous affinity, their peristome with a 2:1 formula most likely results from reduction, through the loss of the asymmetric first-late division. The asymmetric division of the ciliate mosses may not be homologous to that of the haplolepideae (Shaw, Mishler, and Anderson, 1989). Assuming that the genetic complexity behind the asymmetric division in both groups is similar, we cannot exclude the possibility that a loss of it or a reversal to a symmetric division may be possible among ciliate mosses and may even occur in such genera as Mielichhoferia with a peristome formula of 4:2:4 (Shaw and Crum, 1984; Shaw and Rohrer, 1984). Since the possibility of a reversal to a symmetric division cannot be excluded, a bryalean origin of the Orthotrichaceae will need to be further examined in comparison with a broad sample of ciliate taxa.

Hedenäs (1994) considered Schlotheimia to be pleurocarpous and, based on this interpretation, suggested that the Orthotrichaceae «are rather close to the clade where most pleurocarpous mosses belong.» He further hypothesized that «the transition to pleurocarpy must have been gradual,» and in this scenario the Orthotrichaceae would occupy an intermediate position. The Macromitrioideae differ, however, from typical pleurocarps by several characters and should rather be considered cladocarpous (La Farge-England, 1996). In genera where both cladocarpy and acrocarpy occur, the former seems to be restricted to terminal taxa, suggesting that the trend is from acrocarpy to cladocarpy with no obvious cases of reversals (La Farge-England, 1996). Such a general trend may suggest that in the Orthotrichaceae too, the primitive condition is acrocarpy. Thus if the Orthotrichales are indeed reduced ciliate mosses, their putative sister group would most likely belong to a group of acrocarpous Eubryales.

Subfamilial phylogenetic relationships
The ten orthotrichaceous genera remaining in the present analysis are distributed among two clades that are moderately to strongly supported by either method of analysis, with bootstrap values (BV) ranging from 77 to 94% and decay indexes (DI) of 2 and 3. The orthotrichoid clade combines the Zygodontoideae (Zygodon sections Zygodon and Bryoides) and the Orthotrichoideae (Bryodixonia, Orthotrichum, and Ulota), while the macromitrioid clade includes all the Macromitrioideae (Cardotiella, Desmotheca, Groutiella, Macrocoma, Macromitrium, and Schlotheimia) as well as Zygodon obtusifolius (Zygodon sect. Obtusifolii). Zygodon obtusifolius has retained several characters considered here plesiomorphic, such as acrocarpy, smooth, cucullate calyptrae, but exhibits also some derived features reminiscent of some Macromitrioideae (strongly bulging cells, coarse papillae, undifferentiated basal cells, etc.; see Goffinet, 1997c). If Z. obtusifolius is to be retained within the Zygodontoideae, the unexpected relationship with the Macromitrioideae as proposed by rbcL sequence data may be an artifact due to a long branch attraction (Hendy and Penny, 1989) or an indication of hybridization involving chloroplast capture from a macromitrioid taxon (Soltis and Kuzoff, 1995). In the latter case, the topology obtained here may represent the «correct» gene tree, but deviates from the true phylogeny of the taxa (Doyle, 1992). This hypothesis is currently being tested by comparing sequence data of the nuclear gene 18S.

With the exclusion of the Drummondioideae, Vitt's (1982a) phylogenetic arrangement of subfamilies would have the Zygodontoideae (cucullate calyptrae) basal to a dichotomy between the Orthotrichoideae and the Macromitrioideae (both typically or exclusively have large mitrate calyptrae). The sister group relationship between the orthotrichoid and the macromitrioid clades proposed here deviates from Vitt's (1982a) phylogenetic concept of the family by the inclusion of the Zygodontoideae in the former clade, and parsimony would need to be relaxed by six steps for his concept to be satisfied (without constraining the affinities of Z. obtusifolius). The monophyly of Zygodon and thus Zygodontoideae is not supported by rbcL sequence data either, and the relationship of sections Zygodon and Bryoides to the Orthotrichoideae sensu stricto remain unresolved. If Zygodon is indeed paraphyletic, section Bryoides would most likely be the most primitive taxon, given that it shares the plesiomorphic state of «smooth laminal cells» with either a splachnaceous or a typical bryaceous ancestor. The two sections of Zygodon differ on average by 40 mutations, while only 28 changes separate section Zygodon from the Orthotrichoideae, compared to 43 changes between section Bryoides and the Orthotrichoideae. Such divergences may be indicative of the monophyly of a clade composed of section Zygodon and the Orthotrichoideae and support a single origin of papillae that would define this clade.

With regards to the Macromitrioideae, both methods of analysis yield two strongly supported monophyletic clades (Figs. 2, 4): one composed of all three species of Schlotheimia, the other including all remaining Macromitrioideae (i.e., Cardotiella, Desmotheca, Groutiella, Macrocoma, and Macromitrium). Vitt, Koponen, and Norris (1993) suggested that Schlotheimia and Cardotiella should be placed in a separate subfamily based on the smooth and distinctly lobate calyptra. RbcL data do not support such relationship as ten more steps are needed to unite Schlotheimia and Cardotiella into a monophyletic clade. The species of Schlotheimia differ, on average, from other Macromitrioideae (including Cardotiella) by 47 bases. By contrast, the average distance between these other macromitrioid genera is only 23 bases. If rates of molecular evolution are assumed to be similar among these taxa (a reasonable assumption considering both clades are predominantly phyllodioicous, and epiphytic in tropical montane forests; see Britten, 1986), differences in nucleotide sequences may indicate a relatively ancient divergence between Schlotheimia and the other Macromitrioideae. Among the morphological characters that separate Schlotheimia from other Macromitrioideae, those related to the calyptrae (shape, anatomy, outline) are shared only with Cardotiella (Vitt, 1981b), whereas the remaining ones (cell areolation, color of leaves) are plesiomorphic or autapomorphic (Goffinet, 1997c) and thus phylogenetically uninformative. The pleisiomorphic state of «smooth laminal cells» may, as was argued in the case of Zygodon sect. Bryoides, be indicative of the primitive position of Schlotheimia in the evolution of the Macromitrioideae. Considering both the morphological and molecular divergences between Schlotheimia and the remaining Macromitrioideae, and the strong support for the monophyly of both clades on molecular grounds, accommodating Schlotheimia in its own subfamily may better reflect the evolutionary relationship among these clades.

Brotherus (1925, as Pseudo-Macromitrioideae), Walther (1983), and Crum (1987) isolated Desmotheca from other Macromitrioideae on the basis of its dimorphic sterile and fertile branches. Vitt (1990), however, argued that excluding Desmotheca from the Macromitrioideae would most certainly result in the paraphyly of the later subfamily. Based on variation in the rbcL sequences, the phylogenetic affinities of Desmotheca clearly lay with Macromitrium sensu lato, but remain ambiguous with regard to its sister group within this clade. Whether Desmotheca should be retained in its own subfamily, and placed sister to the Macromitrioideae (excluding Schlotheimia), or be inserted within the latter, is not clear. In 33 MPTs (including the 15 that share a monophyletic Orthotrichaceae) Desmotheca occupies a derived position (Fig. 3), nested between two Macromitrium species. In the remaining six MPTs as well as in the NJ tree (Figs. 3, 4), Desmotheca is sister to the Macromitrioideae (Schlotheimia excluded). The phenetic distance between Desmotheca and other macromitrioid genera (Table 4; excluding Schlotheimia) are, on average, similar to those between Orthotrichum and Ulota (20 bases), in the Orthotrichoideae, suggesting that placing Desmotheca in a distinct subfamily may not be appropriate.

Phylogenetic relationships of Orthotrichaceous genera
Analysis of rbcL sequence using either the MP or the NJ method suggests that four of the five larger genera, namely Macromitrium, Orthotrichum, Ulota, and Zygodon, are not monophyletic. The paraphyly of Zygodon has been addressed earlier. Bryodixonia is a monotypic genus endemic to New Zealand. Sainsbury (1945) argued for a generic distinction from Ulota on the basis of «the highly differentiated and conspicuous perichaetial bracts and the diminutive calyptra» and the immersed capsule. In addition, the perichaetial leaves have prorate basal laminal cells (Goffinet, 1997c), a feature otherwise unknown from the Orthotrichoideae. Bryodixonia does, however, share many of the characters found in the genus Ulota, such as very thick-walled cauline cells, the differentiated marginal cells of the lamina, and the flexuose to crisped leaves. Bryodixonia may thus be patristically very derived; however, in cladistic terms it may not deserve taxonomic recognition at the generic level. RbcL sequences of Ulota lutea and Bryodixonia perichaetialis differ only by six mutations, a degree of divergence that is similar to that found between species of Groutiella (4) or Schlotheimia (8–16), but, moreover, it is less than the divergence recorded between the two species of Ulota (14; Table 4). Immersed capsules are characteristic of many mosses that are taxonomically unrelated but share a xerophytic habitat (Vitt, 1981a). Within the Orthotrichaceae completely immersed capsules are characteristic for Schlotheimia sect. Stegotheca, and a shortening of the setae leading ultimately to an immersed capsule occurs in Orthotrichum subg. Gymnoporus sect. Leiocarpa, subg. Pulchella sect. Rivularia, and sect. Diaphana, as well as in subg. Orthotrichum (Lewinsky, 1993). Considering the low degree of morphological and molecular divergence of Bryodixonia, segregation at the generic level does not seem appropriate.

The monophyly of Ulota (including B. perichaetialis) is compromised by O. lyellii in the NJ tree and in 13 MPT (in 13 other MPTs their relationship is not resolved). Strong affinities of O. lyellii for Ulota (MP: 99% BV and DI of 5) and U. obtusiuscula in particular (13 MPTs) may indicate that among the different lineages of Orthotrichum, subg. Gymnoporus (Braithw.) Limpr. is the most closely related to Ulota. Among the characters that subg. Gymnoporus sect. Leiocarpa (see description in Lewinsky, 1993) shares with Ulota, only the long flexuose vegetative leaves may be derived within the Orthotrichoideae and thus be indicative of common ancestry. The genus Ulota (even if including Bryodixonia) is morphologically well defined from Orthotrichum, suggesting that the paraphyly of Orthotrichum, if confirmed, would need to be resolved by dividing the genus into discrete entities rather than broadening the concept of Orthotrichum by including Ulota. The relationships of the subg. Orthophyllum Delogn. (O. obtusifolium) and subg. Orthotrichum (O. anomalum) are not unambiguously resolved either: they form sister taxa in 13 MPTs as well as the NJ tree, while in 13 other MPTs, O. obtusifolium is sister to a clade composed of the remaining Orthotrichoideae. Orthotrichum obtusifolium had been placed together with the related O. gymnostomum Brid. in the genus Nyholmiella (see Lewinsky, 1993, for history), based on «the obtuse leaves with plane or incurved leaf margin and incrassate leaf-cells with a stout central papillae on each side» (Lewinsky, 1993). Patterns in peristome ornamentation are similar to those observed elsewhere in the genus (Lewinsky, 1993), and it may therefore be more parsimonious to retain subg. Orthophyllum in Orthotrichum.

Within the macromitrioid clade (Fig. 4) the relationships remain ambiguous too (except for Schlotheimia see above), either because Macromitrium and Macrocoma truly are paraphyletic, or because of an insufficient taxon sample. The genus Macromitrium is, with over 250 species, by far the most speciose genus of the Orthotrichaceae (Vitt, 1982a). Mitten (1869) divided the genus in four sections, to which Buck (1991) recently added sect. Reverberatum. Two of these have recently been raised to the genus level, namely Micromitrium (now Groutiella Steere) and Macrocoma Grout. The genus Macromitrium remains, however, morphologically extremely diverse in terms of size of the plant, degree of differentiation of the basal cells, shape of the urn, and laminal cell shape, orientation, and ornamentation. The relatively high cost in terms of parsimony, for a monophyletic Macromitrium (nine steps) may be seen as just one other indication that the genus as currently defined is still a heterogeneous assemblage. Groutiella differs from Macromitrium by the marginal limbidium of hyaline elongate cells and a short calyptra covering only the upper portion of the urn. Except for G. tomentosa, Groutiella is restricted to the Neotropics, where ten species occur. The sister species in all shortest trees is Macromitrium longifolium, a neotropical endemic, rather than any of two paleotropical taxa (M. richardii is known from Africa and the Americas [van Rooy and van Wyk, 1992; Vitt 1993] and belongs to M. ligulare group of the Old World; M. incurvifolium occurs throughout the Pacific Ocean [Vitt and Ramsay, 1985]). Whether these putative affinities of Groutiella for neotropical Macromitria indicate a common ancestry with a distinct neotropical lineage of Macromitrium needs to be further investigated.

The genus Macrocoma is composed of two subgenera, subg. Trachyphyllum (M. papillosa) and subg. Macrocoma (M. tenuis), that differ by a series of characters, but particularly by the well-developed peristome of the former (Vitt, 1980). Our molecular data suggest that these two taxa, too, form an artificial group; three additional steps are needed to restore the monophyly of Macrocoma. Macrocoma papillosa is found in a basal position among macromitrioid taxa (excluding Schlotheimia) in 33 MPTs (Fig. 4), while in the remaining six it is found in a more derived position with both species of Groutiella, Macromitrium longifolium, and Macrocoma tenuis. While the Macromitrioideae are typically cladocarpous (i.e., with their perichaetia terminal on lateral branches) and have dimorphic leaves between stem and branches, M. papillosa is clado- and acrocarpous (i.e, with perichaetia terminal on lateral branches and on the stem) and the leaves are not differentiated into stem and branch leaves (Goffinet, 1997c). True acrocarpy also occurs in subg. Macrocoma (e.g., M. braziliensis [Mitt] Vitt), while other taxa of this subgenus are strictly cladocarpous (e.g., M. tenuis [Hook. & Grev.] Vitt). Acrocarpy and cladocarpy have not been reported before from the same taxon, not to mention from the same individual (see La Farge-England, 1996). The combination of plesiomorphies such as undifferentiated stem and branch leaves, terminal cauline gametangia, and the complete double peristome may be a strong indication that subg. Trachyphyllum is a primitive clade not only when compared to subg. Macrocoma (Vitt, 1982c), but maybe even with regard to the evolution of the Macromitrioideae.

Affinities of excluded taxa
Critically addressing the relationships of the taxa here excluded from the Orthotrichaceae is beyond the scope of the present study and would need a broader sampling of haplolepideous taxa. A suite of unique characters combined with the lack of characters that are phylogenetically crucial may always hamper determining sister group relationships based on morphology only. Crum (1987) already suggested that gametophytic characters may not suffice to resolve the phylogenetic relationship of Drummondia. The same opinion prevails with regards to Amphidium. Brotherus (1925), Anderson and Crum (1958), and Vitt (1973, 1982b, 1984) placed Amphidium near Rhabdoweisia. Our results, though preliminary, do not suggest that these genera are closely related, and future study may need to consider alternative relationships, as for example with Glyphomitrium, due to the overall similarity of Amphidium lapponicum-Glyphomitrium daviesii.

Presence of a peristome allows for a more explicit hypothesis to be made regarding the systematic position of the Erpodiaceae and the Rhachitheciaceae. Edwards (1979), as part of a review of the haplolepideous peristome, examined the peristome architecture of Venturiella and concluded that the teeth are «strongly dorsally trabeculate, and also have a rudimentary unthickened basal exostome,» and that «these characters are of a haplolepideous peristome although not of the dicranaceous type» but of a distinct type, the Seligeria type. This peristome type is characterized by little ventral thickening, strong dorsal trabeculae, and an exostome reduced to a thin, smooth membrane adhering to the trabeculae. This combination of characters has also been observed in the Rhachitheciaceae and has been interpreted as a possible indication of haplolepideous affinities of the family (Goffinet, 1997a). Molecular data thus tend to confirm these hypotheses, and consequently the peristome of the Rhachitheciaceae and the Erpodiaceae should be regarded as derived, through reduction, from a typical haplolepideous peristome. Though the monophyly of a group of taxa sharing the Seligeria-type peristome has not been critically examined, the nearly identical peristomes of Rhachithecium, Glyphomitrium (Ptychomitriaceae), and Blindia (Seligeriaceae) may be seen as an indication of close phylogenetic relationships, despite gametophytic differences. Alternatively, the Rhachitheciaceae may be more closely related to Rhabdoweisia in the Dicranales, considering the similarities in the overall habit, leaf shape, and cell shape and differentiation.

Phylogenetic conclusions
Sequence data of the chloroplast gene rbcL are useful in circumscribing the Orthotrichales, particularly with regard to the systematic position of taxa lacking peristome features that are central to the classification of mosses. Analyses of the variation in the nucleotide sequence using either the parsimony or the distance method strongly suggest that the Orthotrichales are polyphyletic and that the Erpodiaceae and the Rhachitheciaceae are of haplolepideous affinities as suggested by their Seligeria-type peristome. The Orthotrichaceae too, are shown to be an artificial assemblage due to the current inclusion of Amphidium and Drummondia, two genera better placed among the Haplolepideae. The Orthotrichaceae are only distantly related to the latter clade and should rather be considered a member of a derived diplolepideous clade. The relationship to the Splachnales and the ciliate mosses remains unsettled, but at present all three lineages are best considered monophyletic. Molecular data do not support Vitt's (1982a) subfamilial phylogeny and instead suggest that the Orthotrichaceae are composed of two lineages. The first consists of the Zygodontoideae and the Orthotrichoideae, while the second includes all Macromitrioideae. Putative basal taxa within these clades may be represented by Zygodon sect. Bryoides and Schlotheimia, respectively, which are characterized by smooth laminal cells.

The Orthotrichaceae are now composed of 20 genera: Bryodixonia, Cardotiella, Ceuthotheca, Desmotheca, Florschuetziella, Groutiella, Leiomitrium, Leptodontiopsis, Leratia, Macrocoma, Macromitrium, Muelleriella, Orthomitrium, Orthotrichum, Pleurorthotrichum, Schlotheimia, Stenomitrium, Stoneobryum, Ulota, and Zygodon. Examination of cladistic relationships and associated phenetic distances suggests that the monotypic genus Bryodixonia may be better regarded as a patristically derived species of Ulota. Our molecular data furthermore reveal that larger genera such as Macromitrium and Zygodon may merely represent evolutionary grades. Gene data obviously have made a significant contribution in resolving the circumscription of the Orthotrichaceae and the relationship of the main lineages. Above all, however, this molecular study has laid the foundation for critically reexamining the morphological characters that are central to the generic concept used in the Orthotrichaceae.

	FOOTNOTES

 
¹ This study is part of a Ph.D. dissertation presented by the senior author to the Faculty of Graduate Studies of the University of Alberta. This study was supported by Grants A-3797 and A-6390 from the Nature Sciences and Engineering Research Council of Canada to RJB and DHV, respectively. Field work was made possible through a grant by the National Geographic Society to DHV and BG. We are grateful to William Buck (NY), Celina Matteri (BA), and Heinar Streimann (CANB) for kindly sending us fresh material of Cardotiella quinquefaria, Macrocoma papillosa, and Desmotheca apiculata, respectively, as well as Ross Hastings (PMAE) for the loan of Ptychomitrium gardneri for DNA extraction. Finally we thank G. Zurawski for the set of sequencing primers.

² Author for correspondence, current address: Department of Botany, Box 90339, Duke University, Durham, NC 27708.

³ Present address: CSIRO, Division of Plant Industry, Australian National Herbarium, Centre for Plant Biodiversity Research, Molecular Systematics Lab, GPO Box 1600, Canberra, ACT, 2601, Australia.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Anderson, L. E., and H. Crum. 1958 Cytotaxonomic studies on mosses of the Canadian Rocky Mountains. Bulletin of the National Museum of Canada 160: 1–89.

Andersson, I., S. Knight, G. Schneider, Y. Lindqvist, T. Lundqvist, C.-I. Bränden, and G. H. Lorimer. 1989 Crystal structure of the active site of ribulose-bisphosphate carboxylase. Nature 337: 229–234. [CrossRef]

Barker, N. P., H. P. Linder, and E. H. Harley. 1995 Polyphyly of Arundinoideae (Poaceae): evidence from rbcL sequence data. Systematic Botany 20: 423–435. [CrossRef][Web of Science]

Baum, D. A., K. J. Sytsma, and P. C. Hoch. 1994 A phylogenetic analysis of Epilobium (Onagraceae) based on nuclear ribosomal DNA sequences. Systematic Botany 19: 363–388. [CrossRef][Web of Science]

Blomquist, H. L., and L. L. Robertson. 1941 The development of the peristome in Aulacomnium heterostichum. Bulletin of the Torrey Botanical Club 68: 569–584. [CrossRef]

Bopp, M., and I. Capesius. 1996 New aspects of Bryophyte taxonomy provided by a molecular approach. Botanica Acta 109: 368–372. [Web of Science]

Bremer, K. 1988. The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 46: 1801–1817. [CrossRef]

Britten, R. J. 1986 Rates of DNA sequence evolution differ between taxonomic groups. Science 231: 1393–1398. [Abstract/Free Full Text]

Brotherus, V. F. 1909 Musci (Laubmoose). III. Unterklasse Bryales: II Spezieller Teil. In A. Engler's and K. Prantl [eds.], Die Natüurlichen Pflanzenfamilien 1 (3, I, II), 277–1246. W. Engelmann, Leipzig.

———. 1925 Musci (Laubmoose) III. Unterklasse Bryales: II Spezieller Teil. In A. Engler's and K. Prantl [eds.], Die Natüurlichen Pflanzenfamilien, 2d ed., vol. 11, 1–542. W. Engelmann, Leipzig.

Buck, W. R. 1991 Contributions to the moss flora of Guyana. Memoirs of the New Botanical Garden 64: 184–196.

———, and D. H. Vitt. 1986 Suggestions for a new familial classification of pleurocarpous mosses. Taxon 35: 21–60.

Capesius, I. 1995 A molecular phylogeny of Bryophytes based on the nuclear encoded 18S rRNA genes. Journal of Plant Physiology 146: 59–63. [Web of Science]

Chase, M. W., et al. 1993 Phylogenetics of seed plants: an analysis of nucleotide sequences from the plastid gene rbcL. Annals of the Missouri Botanical Garden 80: 528–580. [CrossRef][Web of Science]

Churchill, S. P., and E. L. Linares C. 1995 Prodromus Bryologiae Novo-Granatensis. Introduccion a la Flora de Musgos de Colombia. Parte II. Bibliotheca «Jose Jeronimo Triana» 12: 455–924.

Conti, E., A. Fischbach, and K. J. Sytsma. 1993 Tribal relationships in Onagraceae: Implications from rbcL sequence data. Annals of the Missouri Botanical Garden 80: 672–685. [CrossRef][Web of Science]

Crosby, M. R. 1980 The diversity and relationships of mosses. In R. J. Taylor and A. E. Leviton [eds.], The Mosses of North America, 115–129. California Academy of Science, San Francisco, CA.

Crum, H. 1972 A taxonomic account of the Erpodiaceae. Nova Hedwigia 23: 201–224.

———. 1987 Leratia and its place in the Orthotrichaceae, subfamily Macromitrioideae. Memoirs of the New York Botanical Garden 45: 602–606.

De Luna, E. 1995 The circumscription and phylogenetic relationships of the Hedwigiaceae (Musci). Systematic Botany 20: 347–373.

Dixon, H. N. 1932 Classification of mosses. In F. Verdoorn [ed.], Manual of bryology, 397–412. Martius Nijhoff, The Hague.

Donoghue, M. J., R. G. Olmstead, J. F. Smith, and J. D. Palmer. 1992 Phylogenetic relationships of Dipsacales based on rbcL sequences. Annals of the Missouri Botanical Garden 79: 333–345.

Doyle, J. J. 1992 Gene trees and species trees: molecular systematics as one-character taxonomy. Systematic Botany 17: 144–163. [CrossRef][Web of Science]

———, and J. L. Doyle. 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15.

During, H. J. 1977 A taxonomical revision of the Garovaglioideae (Pterobryaceae, Musci). Bryophytorum Bibliotheca 12: 1–244.

Edwards, S. R. 1979 Taxonomic implications of cell patterns in haplolepidous moss peristomes. In G. S. C. Clarke and J. G. Duckett [eds.], Bryophyte systematics, 317–346. Academic Press, London.

———. 1984 Homologies and inter-relations of moss peristomes. In R. M. Schuster [ed.], New manual of bryology, vol. 2, 658–695. Hattori Botanical Laboratory, Nichinan.

Farris, J. S. 1972 Estimating phylogenetic trees from distance matrix. American Naturalist 106: 645–668. [CrossRef][Web of Science]

Felsenstein, J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791. [CrossRef][Web of Science]

Fitch, W. M. 1971 Toward defining the course of evolution: minimum change for a specific tree topology. Systematic Zoology 20: 406–416. [CrossRef][Web of Science]

Fleischer, M. 1920 Natürlisches System der Laubmoose. Hedwigia 61: 390–400.

Frey, W. 1977 Neue Vorstellungen über die Verwandschaftengruppen und die Stammesgeschichte der Laubmoose. In W. Frey, H. Hurka, and F. Obervwinkler [eds.], Beiträge zur Biologie der niederen Pflanzen, 117–139. Stuttgart.

———. 1990 Genoelemente prä-angiospermen Ursprungs bei Bryophyten. Botanischer Jahresbericht der Systematik 111: 433–456.

———, J.-P. Frahm, E. Fischer, and W. Lobin. 1995 Die Moos- und Farnpflanzen Europas. Kleine Kryptogamenflora Band IV: XII + 426 p. Gustav Fischer Verlag, Stuttgart.

Goffinet, B. 1997a (1998) The Rhachitheciaceae: revised circumscription and ordinal affinities. Bryologist 100: 425–439. [Web of Science]

———. 1997b A reconsideration of the affinities of Kleioweisiopsis, Pleurozygodontopsis, Trigonodictyon, and the Microtheciellaceae (Bryopsida, Orthotrichales). Journal of Bryology, in press.

———. 1997c Circumscription and phylogeny of the Orthotrichales. Ph.D. dissertation. University of Alberta, Edmonton, Canada.

Hasebe, M., P. G. Wolf, K. M. Pryer, K. Ueda, M. Ito, R. Sano, G. J. Gastony, J. Yokoyama, J. R. Manhart, N. Murakami, E. H. Crane, C. H. Haufler, and W. D. Hauk. 1995 Fern phylogeny based on rbcL nucleotide sequences. American Fern Journal 85: 134–181. [CrossRef]

Haufler, C. H., and T. A. Ranker. 1995 RbcL sequences provide phylogenetic insights among sister species of the fern genus Polypodium. American Fern Journal 85: 361–374. [CrossRef][Web of Science]

Hedderson, T. A., R. C. Chapman, and W. R. Rootes. 1996 Phylogenetic relationships of bryophytes inferred from nuclear-encoded rRNA gene sequences. Plant Systematic and Evolution 200: 213–224. [CrossRef]

———, C. J. Cox, and J. G. Gibbings. 1997 Phylogenetic relationships of the Wardiaceae (Musci) inferred from 18s rRNA and rps4 gene sequences. American Journal of Botany 84 (Supplement): 25.

Hedenäs, L. 1994 The basal pleurocarpous diplolepideous mosses—a cladistic approach. Bryologist 97: 225–243. [CrossRef][Web of Science]

Hendy, M. D., and D. Penny. 1989 A framework for the quantitative study of evolutionary trees. Systematic Zoology 38: 297–309. [CrossRef][Web of Science]

Hillis, D. M., and J. P. Huelsenbeck. 1992 Signal, noise, and reliability in molecular phylogenetic analyses. Journal of Heredity 83: 189–195. [Abstract/Free Full Text]

———, and J. J.Bull. 1993 An empirical test for bootstrapping as a method for assessing confidence in phylogenetic analysis. Systematic Biology 42: 182–192. [CrossRef][Web of Science]

Källersjö, M., J. S. Farris, A. G. Kluge, and C. Bult. 1992 Skewness and permutation. Cladistics 8: 275–287.

Käss, E., and M. Wink. 1995 Molecular phylogeny of the Papilionoideae (family Leguminosae): rbcL gene sequences versus chemical taxonomy. Botanica Acta 108: 149–162.

Kellog, E. A., and N. D. Juliano. 1997 The structure and function of RuBisCo and their implications for systematic studies. American Journal of Botany 84: 413–428. [Abstract]

Kim, J., F. J. Rohlf, and R. R. Sokal. 1993 The accuracy of phylogenetic estimation using the neighbor-joining method. Evolution 47: 471–486. [CrossRef][Web of Science]

Koponen, A. 1977 The peristome and spores in Splachnaceae and theirs evolutionary and systematic significance. Bryophytorum Bibliotheca 13: 535–567.

———. 1982 On the structure and function of the peristome in Splachnaceae. Journal of the Hattori Botanical Laboratory 53: 73–98.

———. 1983 Studies on the generic concept in the classification of the moss family Splachnaceae. Publication from the Department of Botany, University of Helsinki 11: 1–48.

Kumar, S., K. Tamura, and M. Nei. 1993 MEGA: Molecular Evolutionary Genetic Analysis, version 1.0. Computer program distributed by Pennsylvania State University, University Park, PA.

La Farge-England, C. 1996 Growth-form, branching pattern, and perichaetial position in mosses: cladocarpy and pleurocarpy redefined. Bryologist 99: 170–186. [CrossRef][Web of Science]

Lewinsky, J. 1976 On the systematic position of Amphidium Schimp. Lindbergia 3: 227–231.

———. 1989 Does the orthotrichaceous type of peristome exist? A study of peristome evolution in the genus Orthotrichum Hedw. with possible derivation of the haplolepideous peristome. Journal of the Hattori Botanical Laboratory 67: 335–363.

———. 1990 Zygodon Hook. & Tayl. in Australasia: a taxonomic revision including SEM studies of peristomes. Lindbergia 15: 109–139.

———. 1993 A synopsis of the genus Orthotrichum Hedw. (Musci, Orthotrichaceae). Bryobrothera 2: 1–59.

Lewinsky-Haapasaari, J., and M. R. Crosby. 1996 Orthomitrium tuberculatum (Orthotrichaceae), a new genus and species from Guizhou, China. Novon 6: 1–5.

Maddison, D. R. 1991 The discovery and importance of multiple islands of most-parsimonious trees. Systematic Zoology 40: 315–328. [CrossRef]

Malta, N. 1926 Die Gattung Zygodon Hook. et Tayl. Eine monographische Studie. Latvijas Universitates Botaniská Dárza Darbi 1: 1–184.

Manhart, J. R. 1994 Phylogenetic analysis of green plant rbcL sequences. Molecular Phylogenetics and Evolution 3: 114–127. [CrossRef][Medline]

Meusel, H. 1935 Wuchsformen und Wuchstypen der europäischen Laubmoose. Nova Acta Leopoldina Neue Folge 3: 123–277.

Mishler, B. D., L. A. Lewis, M. A. Buchheim, K. S. Renzaglia, D. J. Garbary, C. F. Delwiche, F. W. Zechman, T. S. Kantz, and R. L. Chapman. 1994 Phylogenetic relationships of the «green algae» and «bryophyte.» Annals of the Missouri Botanical Garden 81: 451–483. [CrossRef][Web of Science]

———, P. H. Thrall, J. S. Hopple, Jr., E. De Luna, and R. Vilgalys. 1992 A molecular approach to the phylogeny of bryophytes: cladistic analysis of chloroplast-encoded 16S and 23S ribosomal RNA genes. Bryologist 95: 172–180. [CrossRef][Web of Science]

Mitten, W. 1869 Musci Austro-americani. Journal of the Linnean Society of Botany 12: 1–659.

Morgan, D. R., and D. E. Soltis. 1993 Phylogenetic relationships among members of Saxifragaceae sensu lato based on rbcL sequence data. Annals of the Missouri Botanical Garden 80: 631–660. [CrossRef][Web of Science]

Nishimura, N., and K. Watanabe. 1992 SEM observations on the peristome teeth of mosses, 2. Isobryales. Natural Environmental Science Research 5: 83–101.

Olmstead, R. G., and J. A. Sweere. 1994 Combining data in phylogenetic systematics: An empirical approach using three molecular data sets in the Solanaceae. Systematic Biology 43: 467–481. [CrossRef][Web of Science]

Pryer, K. M., A. R. Smith, and J. E. Skog. 1995 Phylogenetic relationships of extant ferns based on evidence from morphology and rbcL sequences. American Fern Journal 85: 205–282. [CrossRef][Web of Science]

Reimers, H. 1954 Bryophyta. In H. Melchior and E. Wedermann [eds.], A. Engler's Syllabus der Pflanzenfamilien, 12th ed., 218–268. Gebrürder Bornträgger, Berlin.

Robinson, H. 1971 A revised classification for the orders and families of mosses. Phytologia 21: 289–293.

Russo, C. A. M., N. Takezaki, and M. Nei. 1996 Efficiencies of different genes and different tree-building methods in recovering a known vertebrate phylogeny. Molecular Biology and Evolution 13: 525–536. [Abstract]

Sainsbury, G. O. K. 1945 New and critical species of New Zealand mosses. Transaction of the Royal Society of New Zealand 75: 169–185.

Saitou, N., and M. Nei. 1987 The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4: 406–425. [Abstract]

Sang, T. 1995 New measurements of distribution of homoplasy and reliability of parsimonious cladograms. Taxon 44: 77–82.

Sanger, F., S. Nicklen, and A. R. Coulson. 1977 DNA sequencing with chain-terminating inhibitors. Proceedings of the National Academy of Sciences, USA 74: 5463–5467. [Abstract/Free Full Text]

Schwartz, O. M. 1994 The development of the peristome-forming layers in the Funariaceae. International Journal of Plant Sciences 155: 640–657. [CrossRef]

Shaw, J. 1985 Peristome structure in the Orthotrichaceae. Journal of the Hattori Botanical Laboratory 60: 119–136.

———, L. E. Anderson, and B. D. Mishler. 1989 Peristome development in mosses in relation to systematics and evolution. III. Funaria hygrometrica, Bryum pseudocapillare, and B. bicolor. Systematic Botany 14: 24–36.

———, and H. Crum. 1984 Peristome homology in Mielichhoferia and a taxonomic account of North American species. Journal of the Hattori Botanical Laboratory 57: 363–381.

———, B. D. Mishler, and L. E. Anderson. 1989 Peristome development in mosses in relation to systematics and evolution. IV. Haplolepidae: Ditrichaceae and Dicranaceae. Bryologist 92: 314–325. [CrossRef][Web of Science]

———, and J. R. Rohrer. 1984 Endostomial architecture in diplolepideous mosses. Journal of the Hattori Botanical Laboratory 57: 41–61.

Soltis, D. E., and R. K. Kuzoff. 1995 Discordance between nuclear and chloroplast phylogeneis in the Heuchera group (Saxifragaceae). Evolution 49: 727–742. [CrossRef][Web of Science]

Swofford, D. L. 1993 PAUP: phylogenetic analysis using parsimony, version 3.1.1. Computer Program distributed by the Illinois Natural History Survey, Champaign, IL.

van Rooy, J., and A. E. van Wyk. 1992 A conspectus of the subfamily Macromitrioideae (Bryopsida: Orthotrichaceae) in Southern Africa. Bryologist 95: 205–215. [CrossRef][Web of Science]

Vitt, D. H. 1970 The family Orthotrichaceae (Musci) in North America, north of Mexico. Ph.D. dissertation. University of Michigan, Ann Arbor, MI.

———. 1972 A monograph of the genus Drummondia. Canadian Journal of Botany 50: 1191–1208. [Web of Science]

———. 1973 A revision of the genus Orthotrichum in North America, north of Mexico. Bryophytorum Bibliotheca 1: 1–208.

———. 1980 (1981) The genus Macrocoma I. Typification of names and taxonomy of the species. Bryologist 83: 405–436. [CrossRef][Web of Science]

———. 1981a Adaptive modes of the moss sporophyte. Bryologist 84: 166–186. [CrossRef][Web of Science]

———. 1981b The genera Leiomitrium and Cardotiella gen. nov. (Orthotrichaceae). Journal of the Hattori Botanical Laboratory 49: 93–113.

———. 1982a Sphagnopsida and Bryopsida. In S. P. Parker [ed.], Synopsis and classification of living organisms, vol. 1, 307–336. McGraw-Hill, New York, NY.

———. 1982b The genera of the Orthotrichaceae. Beiheft zu Nova Hedwigia 71: 261–268.

———. 1982c Populational variation and speciation in austral mosses. Journal of the Hattori Botanical Laboratory 52: 153–159.

———. 1984 Classification of the Bryopsida. In R. M. Schuster [ed.], New manual of bryology, vol. 2. Hattori Botanical Laboratory, Nichinan, Japan.

———. 1990 Desmotheca (Orthotrichaceae): Gondwanan fragmentation and the origin of a Southeast Asian genus. Tropical Bryology 3: 79–88.

———. 1993 Orthotrichaceae. In A. J. Sharp, H. A. Crum, and P. M. Eckel [eds.], The moss flora of Mexico, 590–657. New York Botanical Garden, New York, NY.

———, B. Goffinet, and T. A. Hedderson. 1997 The ordinal classification of the mosses: Questions and answers for the 1990's. In J. G. Duckett, N. W. Ashton, and J. W. Bates [eds], Bryology for the second century. Leeds: W. S. Mary and the British Bryological Society, in press.

———, T. Koponen, and D. H. Norris. 1993 Bryophyte flora of the Huon Peninsula, Papua New Guinea. LIII. Ulota and Schlotheimia (Orthotrichaceae, Musci). Acta Botanica Fennica: 148: 5–25.

———, and H. P. Ramsay. 1985 The Macromitrium complex in Australasia (Orthotrichaceae: Bryopsida). Part II. Distribution, ecology, and paleogeography. Journal of the Hattori Botanical Laboratory 59: 453–468.

Walther, K. 1983 Bryophytina. Laubmoose. In J. Gerloff and J. Poelt [eds.], A. Engler's Syllabus der Pflanzenfamilien. Aufl. 13, V(2): 1–108. Gebrüder Bornträger, Berlin.

Waters, D. A., M. A. Buchheim, R. A. Dewey, and R. C. Chapman. 1992 Preliminary inferences of the phylogeny of bryophytes from nuclear-encoded ribosomal RNA sequences. American Journal of Botany 79: 459–466. [CrossRef][Web of Science]

Welch, W. H. 1943 The systematic position of the genera Wardia, Hydropogon, and Hydropogonella. Bryologist 46: 25–46.

Wolf, P. G., P. S. Soltis, and D. E. Soltis. 1994 Phylogenetic relationships of Dennstaedtioid ferns: evidence from rbcL sequences. Molecular Phylogenetics and Evolution 3: 383–392. [CrossRef][Medline]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				B. Goffinet, A. J. Shaw, and C. J. Cox Phylogenetic inferences in the dung-moss family Splachnaceae from analyses of cpDNA sequence data and implications for the evolution of entomophily Am. J. Botany, May 1, 2004; 91(5): 748 - 759. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (43) Citing Articles via Google Scholar Google Scholar Articles by Goffinet, B. Articles by Vitt, D. H. Search for Related Content PubMed Articles by Goffinet, B. Articles by Vitt, D. H. Agricola Articles by Goffinet, B. Articles by Vitt, D. H. Social Bookmarking                            What's this?