Molecular Phylogeny of the Upright Erythropeltidales (Compsopogonophyceae, Rhodophyta): Multiple Cryptic Lineages of Erythrotrichia Carnea1

Home , Compsopogonales, Compsopogonophyceae, Erythropeltidales, Erythrotrichia

J. Phycol. 47, 627–637 (2011) 2011 Phycological Society of America DOI: 10.1111/j.1529-8817.2011.00985.x

MOLECULAR PHYLOGENY OF THE UPRIGHT ERYTHROPELTIDALES (COMPSOPOGONOPHYCEAE, RHODOPHYTA): MULTIPLE CRYPTIC LINEAGES OF ERYTHROTRICHIA CARNEA1

Giuseppe C. Zuccarello2 School of Biological Sciences, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand Hwan Su Yoon, HeeJeong Kim Bigelow Laboratory for Ocean Sciences, 180 McKown Point Road, West Boothbay Harbor, Maine 04575, USA Ling Sun School of Biological Sciences, Victoria University of Wellington, P.O. Box 600, Wellington, 6140 New Zealand Susan Loiseaux de Goe¨r 11 rue des Moguerou, 29680 Roscoff, France and John A. West School of Botany, University of Melbourne, Parkville, Victoria 3010, Australia

The phylogeny of morphologically simple algae is Key index words: Compsopogonophyceae; Erythro- problematic due to insufficient morphological char- peltidales; Erythrotrichia; molecular phylogeny; acters to aid in distinguishing species and relation- Porphyrostromium; Rhodophyta; taxonomy ships. The problem is further compounded because multiple evolutionary lineages of morphologically similar species occur in most well-sampled biogeo- The Erythropeltidales are commonly encountered graphic locations; therefore, location cannot be epiphytes in marine environments. The most com- used as a proxy for species. The phylogeny of the mon genera are Erythrotrichia, Erythrocladia, and Sah- upright members of the Erythropeltidales is partially lingia; other genera are less well-known or are clarified by combining molecular data, unialgal cul- geographically restricted (e.g., Porphyrostromium, Smi- ture observations, and worldwide sampling. Our thora); and other genera have only been described results show that there are several well-supported recently (Pseudoerythrocladia, Madagascaria; Zucca- lineages within the Erythropeltidales with only two rello et al. 2010). The usually small stature of these morphologically recognizable taxa at present. The algae, and their frequent co-occurrence on sub- first is the genus Porphyrostromium, with a well-devel- strates, makes their isolation into unialgal culture oped basal crust, which includes two Erythrotrichia essential for accurate observations and molecular species (Porphyrostromium ligulatum comb. nov. and analysis. The ‘‘simple’’ morphology of these algae Porphyrostromium pulvinatum comb. nov.). The sec- led to their placement in one class, the Bangiophy- ond is the branched species Erythrotrichia welwitschii ceae, with other algae of simple morphology (e.g., (Rupr.) Batters. There are also six strongly sup- see van den Hoek et al. 1995). The advent of molec- ported Erythrotrichia carnea–like lineages. While not ular studies has led to the reclassification of the completely satisfactory, we propose that one lineage Bangiophyceae into six classes and a more natural (lineage 2) with samples close to the type locality be taxonomy (Yoon et al. 2006). designated as E. carnea with a specific isolate as an The Erythropeltidales are placed in the class epitype. The lack of morphology to differentiate the Compsopogonophyceae, along with the orders Com- other lineages leads to a taxonomy based solely on psopogonales and Rhodochaetales. In addition to gene sequencing and molecular phylogeny, with rbcL molecular data, this class is supported by some ultra- sequences differentiating the lineages proposed. We structural characters (see Zuccarello et al. 2000, hold off on proposing more species and genera West et al. 2007a,b, Scott et al. 2010). The Compso- until more data and samples can be gathered. pogonophyceae is further characterized by the presence of the low-molecular-weight carbohydrate floridoside, which is also found in the Florideophy- 1Received 11 July 2010. Accepted 15 November 2010. ceae and Porphyridiophyceae (Karsten et al. 1999, 2Author for correspondence: e-mail [email protected]. 2003).

627 628 GIUSEPPE C. ZUCCARELLO ET AL.

The defining features of the Erythropeltidales are in stature, Membranella nitens G. Hollenb. et I. A. Ab- as follows: most species and genera have cells con- bott, has been merged into synonymy with S. naiad- taining a central plastid with a pyrenoid; pit plugs um, again based mostly on molecular characters lacking in all studied species; production of mo- (West and Zuccarello 2009). nospores by the unequal (oblique) division of vege- The remaining two more ubiquitous upright tative cells, forming what is called a lenticular cell, genera in the Erythropeltidales are Erythrotrichia and and the release of this cell as a monospore. The first Porphyrostromium. While there has been much confu- two characters are shared with the Stylonematophy- sion around the application of the names Erythrotrichia, ceae (Zuccarello et al. 2010). Even with this lack of Erythropeltis, Erythrotrichopeltis, and Porphyrostromium many defining characters, the order is well sup- (Wynne 1986, Silva et al. 1996), it appears that the ported by molecular data (Mu¨ller et al. 2001, Yoon correct names presently are Erythrotrichia and et al. 2006). Porphyrostromium. The status of the filamentous genus A sexual cycle, or some features of a sexual cycle, Porphyrostromium Trevis. is not fully resolved, with have been reported in the Compsopogonophyceae. Heerebout (1968) subsuming it into Erythrotrichia Reports of sex are unusual outside the Bangiophy- Aresch. and others not (Kornmann 1984, 1989, ceae (i.e., Bangiales) and Florideophyceae. The Brodie and Irvine 2003, Kikuchi and Shin 2011). At Rhodochaetales was first described as having an present, Porphyrostromium is considered to encompass unusual sexual cycle [in Rhodochaete pulchella Thur. several species with a crustose base (Kornmann 1984, ex Bornet (Magne 1960), as Rhodochaete parvula], 1987), while Erythrotrichia species have a more unicel- although another isolate of R. pulchella is not sexual, lular rhizoidal or discoidal holdfast. Species distinc- but only reproduces by monospores (Zuccarello tions within these two genera have also been et al. 2000). Other reports on sexuality in the controversial, with many species names considered Compsopogonophyceae are few, and evidence of a synonyms. Most characters used to distinguish the sexual cycle is often incomplete (Heerebout 1968, species have been based on substrate, width of the fil- Murray et al. 1972, Kornmann 1984, 1987, Hawkes ament (uni- vs. bi- vs. multiseriate), and attachment 1988, Magne 1990, Nelson 1993, Nelson et al. 2003, structure. Some of these species names (e.g., Erythro- Kikuchi and Shin 2011). trichia australis Levring) have been sunk into a poly- The taxonomy of the crustose members of the typic species E. carnea (Dillwyn) J. Agardh (Guiry and Erythropeltidales was recently investigated using cul- Guiry 2010). Still, little is known about the variability tured isolates and multiple molecular markers and of any of these characters in distinguishing species, wide-range sampling (Zuccarello et al. 2010). This and there has been little comparison of multiple iso- work showed that the diversity of the crustose mem- lates in controlled laboratory conditions. bers is greater than previously suspected. Crustose The Erythropeltidales, while often overlooked in members resembling the genus Erythrocladia were floras due to their small size, are found throughout found in several quite distinct molecular groups that the world. The genera Erythrocladia, Sahlingia, and warranted generic designation. Pseudoerythrocladia Erythrotrichia are especially common and often occur kornmannii J. A. West et Zuccarello was morphologi- as epiphytes on other algae. Worldwide presence of cally distinguishable from Erythrocladia irregularis small organisms may not be due to recent human- Rosenv. by its lack of pyrenoids (Scott et al. 2010), influenced dispersal events, but a characteristic of while the morphology of the genus Madagascaria their small size (e.g., large population sizes, rarity of erythrocladioides J. A. West et N. Kikuchi, a sister group extinction, high probability of dispersal) as stated by to all the other Erythropeltidales, is not easily distin- Bass-Becking (1934, p. 15) and de Wit and Bouvier guished from E. irregularis. The data also showed that (2006, p. 756), ‘‘everything is everywhere but the Porphyropsis coccinea (J. Agardh ex Aresch.) Rosenv. environment selects,’’ and recently expanded by (erect blades without pyrenoids) grouped with others especially for eukaryotes (Fenchel and Finlay the crust P. kornmannii. The data also supported the 2004, Finlay et al. 2004). In the red alga Stylonema genus Sahlingia with bifurcate marginal cells, and this alsidii (Zanardini) K. M. Drew, molecular data indi- genus seemed to be most closely related to the cate that many lineages have worldwide distribu- remaining upright genera (Zuccarello et al. 2010). tions, with some lineages hypothesized to result The upright members of the Erythropeltidales, from recent dispersals (Zuccarello et al. 2008). Also, besides Porphyropsis, have a complex taxonomic his- in E. irregularis, little genetic variation was found in tory. Several large, wide-bladed members in New samples collected from widely dispersed locations Zealand were considered species of Porphyra until (France, the Netherlands, Australia, New Zealand) molecular studies were initiated (Pyrophyllon, Chlido- (Zuccarello et al. 2010). The cause of this wide dis- phyllon; Nelson 1993, Nelson et al. 2003). Others tribution is unknown, but these data suggest that have been known to be distinct from Porphyra. For biogeographic structure is not existent in these example, Smithora naiadum (C. L. Anderson) small organisms, even though resistant long-distance Hollenb, with blades up to 7 cm, is clearly a mem- propagules are unknown. ber of the Erythropeltidales based on molecular This work is a follow-up of our previous publication data (Rintoul et al. 1999). A similar genus, smaller (Zuccarello et al. 2010) on the crustose members of ERYTHROPELTIDALES PHYLOGENY 629 the Erythropeltidales, concentrating on taxonomic Bootstrap support values were calculated with 1,000 replicates issues around upright members of the Erythropelti- to test the stability of monophyletic groups. dales. We specifically posed these questions: (i) Are The data for the Bayesian analyses (in MrBayes v3.0b4; Huelsenbeck and Ronquist 2001) were partitioned by marker present genera supported by molecular data, and are regions and in case of the protein coding genes (psbA, rbcL) new genera warranted? (ii) Can new diagnostic char- additionally by codon position. GTR+I+G was applied to all acters for the genera and species be discovered? (iii) partitions allowing for separate values for each partition. Are there any biogeographic patterns in these organ- Analyses were performed with a random starting tree and two isms? We use mostly unialgal cultures of these organ- runs for 3 · 106 generations, keeping one tree every 100 6 isms, collected from around the world, and use generations. The first 0.5 · 10 generations (burn-in) were discarded, and the remaining 5.0 · 104 trees (representing two multigene analysis to answer these questions. 6 times 2.5 · 10 generations) were used to calculate a posterior probability (PP) at nodes. MATERIALS AND METHODS

Methods for collection, isolation, and maintenance of RESULTS cultures are presented in West and Zuccarello (1999), West (2005), and Zuccarello et al. (2010). Most cultures were Molecular analysis. Our combined three-gene data ) ) maintained in 18C–22C, 5–20 lmol photons Æ m 2 Æ s 1 set (3,038 aligned characters, 661 parsimony-informa- cool-white fluorescent or light-emitting diode lighting at tive sites) produced a mostly well-supported topology 12:12 light:dark daily cycle. Collection information of all taxa (Fig. 1) that was congruent, in all supported used is given in Table S1 (in the supplementary material). Phylogenetic methods. DNA extraction followed a modified branches, with the individual gene trees, each of CTAB method (Zuccarello and Lokhorst 2005). Three genetic which showed poor support for ancestral branch regions were used to determine phylogenetic relationships: (i) orders (Figs. S1–S3 in the supplementary material). Partial nuclear-encoded small ribosomal RNA gene (SSU) The topology was also congruent with previously pub- (between primers G04 and J04) (Saunders and Kraft 1994) lished data on the Erythropeltidales (Zuccarello et al. (800 bp at the 3¢ end of the SSU). This variable region has 2010). In this analysis, M. erythrocladioides formed a been used for species relationships in other red algae (e.g., Broom et al. 1999, Zuccarello et al. 2008). The PCR procedure moderately supported lineage with R. pulchella, and followed Broom et al. (1999). (ii) Plastid-encoded psbA (PSII these were sister to all the other Erythropeltidales. A reaction center protein D1) gene, which was amplified using group of mostly crustose genera (Erythrocladia, Pseud- the primers and conditions described in Yoon et al. (2002). oerythrocladia, and the blade-forming Porphyropsis)was (iii) Plastid-encoded rbcL (LSU of RUBISCO) gene following sister to the remaining samples. All isolates of Sahlin- procedures from Yoon et al. (2002). gia subintegra, a crust with bifurcate marginal cells, All PCR products were electrophoresed in 1% agarose to formed a well-supported clade and were a sister check product size, processed with ExoSAP-IT (USB Corpora- tion, Cleveland, OH, USA), and sequenced commercially group to the remaining Erythropeltidales. (Macrogen Inc., Seoul, Korea). The upright Erythropeltidales are a highly diverse All sequences were compiled in Se-Al version 2.0a.11 assemblage of distinct evolutionary lineages. Several (Rambaut 1996) and aligned by eye. Phylogenetic relationships monophyletic groups were supported. The first based on maximum-parsimony (MP) were inferred with contained samples identified as Porphyrostromium PAUP*4.0b10 (Swofford 2002). We used the Compsopogonales [Porphyrostromium ciliare (Carmich.) M. J. Wynne, [Compsopogon caeruleus (Balb. ex C. Agardh) Mont.] and Rhodochaetales (R. pulchella) as the outgroup to the Erythro- Porphyrostromium boryanum (Mont.) P. C. Silva], peltidales (Yoon et al. 2006, Yokoyama et al. 2009). Trees Porphyrostromium japonicum (Tokida) N. Kikuchi produced from individual gene data sets did not differ in any (Kikuchi and Shin 2011), and a species from Austra- supported branches (>70% bootstrap analysis [BP]) from each lia known as Erythrotrichia ligulata Womersley. This other, so a combined data set was made of all samples for which Porphyrostromium lineage also contained a sample we had all three genes (67 taxa). A partition-homogeneity test collected on Codium fragile subsp. fragile from also found no incongruence between data sets (P = 0.9500, analysis not shown). Oregon (isolate 4758; Fig. 3, i–k) originally identi- MP trees were constructed in PAUP*, using the heuristic fied as Erythrotrichia pulvinata N. L. Gardner, search option, 500 random sequence additions, tree-bisection- although this was only sequenced with the psbA reconstruction branch swapping, unordered and unweighted gene (Fig. S2). The second morphologically identifi- characters, gaps treated as missing data, and 1 million rear- able lineage matched the species E. welwitschii rangements per replicate. Support for individual internal (Rupr.) Batters. This alga was distinguished by its branches was determined by BP (Felsenstein 1985), as imple- uniseriate branched filaments (Fig. 2, h–j, see more mented in PAUP*. For BP, 1,000 bootstrap data sets were generated from resampled data (10 random sequence addi- below) and habitat (midintertidal growing on Ralfsia tions, 1 million rearrangements allowed per replicate). sp. on mollusk shells or stones). The remaining Maximum-likelihood (ML) analyses were performed using algae were nearly all identified initially as E. carnea. RAxML v.7.2.6 (Stamatakis 2006). Tree likelihoods were At least six additional lineages can be designated estimated using 200 replications with site-specific general time based on phylogenetic support and genetic variation reversible (GTR) and site-specific rate heterogeneity (G) model (Fig. 1, lineages 1–6). These lineages were quite dis- for each data partition. For each replication, an independent MP tree from a random starting point was used for the tree tinct, as well defined as generic relationships (see likelihood estimation. Default rapid hill climbing method with variation between Pseudoerythrocladia and Porphyropsis) automatically optimized subtree pruning and regrafting (SPR) and more distinctive than the variation seen rearrangement algorithm was used for best tree search. between species of Porphyrostromium. All of these 630 GIUSEPPE C. ZUCCARELLO ET AL.

4253.1VIC-AUS 1 4252.1 VIC-AUS 4251 VIC-AUS 4398.1 TAS-AUS 4697 SA-AUS 4476 FRA 4388 NCL 88/92/1.0 4272 NSW-AUS 93/100/1.0 4054 GBR 2 4425 VIC-AUS 4056 GBR 4050 IRL 55/81/1.0 4057 GBR 4645 BRA 4675 JPN -/62/1.0 4704 FRA 4469 FRA Erythrotrichia welwitschii 4471.1 FRA 0.06 4474 FRA 86/76/1.0 P. japonicum 4679 JPN 75/96/1.0 P. ligulatum F600 VIC-AUS P. boryanum 4703 FRA Porphyrostromium 80/83/1.0 P. boryanum 4710 FRA P. ciliare 4708 FRA 4672 BRA 3 70/72/1.0 4667 BRA 4058 NSW-AUS 4488 MDG 4482 MDG 65/76/1.0 4274 QLD-AUS 4568 ZAF 5 4396 TAS-AUS 4245 VIC-AUS 98/93/1.0 4247 VIC-AUS 4295 USA 75/96/1.0 4475 NLD 4397 TAS-AUS 6 4473 GBR 4392 TAS-AUS 4398.2 TAS-AUS 4 4393.1 TAS-AUS 3438WA-AUS 4323 IND 4345 IND 3442 USA 4321 MDG 4331 MDG 4320 MDG 4327 MDG Sahlingia subintegra 4269 VIC-AUS F373 NZL 4658 BRA 4262.6b NSW-AUS 4578 ZAF 4464 FRA 4477 FRA Pseudoerythrocladia kornmannii 77/79/1.0 4471.2 FRA 4678 JPN 4073.1 GBR Porphyropsis coccinea 80/89/1.0 4467 FRA 4465 FRA 4394.2 TAS-AUS Erythrocladia irregularis 4393.2 TAS-AUS 4468 NLD -/90/1.0 Madagascaria erythrocladiodes 4480 MDG Rhodochaete pulchella VIC-AUS

FIG. 1. Phylogeny of the Erythropeltidales inferred from maximum-likelihood (ML) analysis ()ln L = 16175.001176) based on three- gene data set (rbcL, SSU, psbA). Support values based on maximum-parsimony (MP) bootstrap ⁄ ML bootstrap ⁄ Bayesian posterior probabilities, respectively. *Values ‡95% bootstrap and ‡0.95 posterior probabilities. Values £50% bootstrap and £0.7 posterior probabilities not shown. JAW isolate number, three-letter country codes shown. Australian states shown (QLD, NSW, SA, TAS, WA) followed by AUS. Erythrotrichia ‘‘carnea’’ lineages numbered. Outgroup Compsopogon caeruleus removed. Scales bar = substitutions per site. lineages contained samples from geographically dis- 4), and samples from Florianopolis, Brazil, isolated tant locations. For example, lineage 6 was sampled on a single host (Codium sp.), two from lineage 3 from Tasmania, Australia (4392, 4397), and Great (4667, 4672) and another from lineage 2 (4645). Britain (4473), and lineage 5 included samples from Relationships between lineages were mainly unre- Australia, USA, the Netherlands, and South Africa. solved. There was a moderately supported grouping This lack of biogeographic structure was even seen of lineages 1 and 2 with E. welwitschii (55 MP BP ⁄ 81 at a local level. Samples from Penguin, Tasmania, ML BP ⁄ 1.0 PP). A sister relationship between lin- Australia, were collected on the same day at the eages 5 and 6 was also well supported in our analy- same site, and yet one was in lineage 5 (4396) and ses (75 MP BP ⁄ 96 ML BP ⁄ 1.0 PP; Fig. 1). the other in lineage 6 (4397). The same was true Several samples showed unusual phylogenetic for isolates on one host (Wollastoniella sp.) from Tin- placements. Isolate 4474 from Brittany, France, did derbox, Tasmania (4398.1, lineage 1; 4398.2, lineage not group with any other sample analyzed, although ERYTHROPELTIDALES PHYLOGENY 631 it was sister, with moderate support, to lineages 1 feature was seen in all lineages except lineage 1, and 2, and E. welwitschii. Isolate 4058 from Sydney, which had a markedly elongate basal cell. Monosp- NSW, Australia was morphologically identified as ores were released through an obvious discharge Erythrotrichia foliiformis South et N. M. Adams due to papilla (Fig. 2t). Lineage 5 was like other lineages its monostromatic blade and rhizoidal base (Fig. 2, except that the tip of the elongate rhizoids had con- k–m). It grouped with uniseriate isolates (e.g., spicuously expanded tips (Fig. 2, v and w), but it 4482) in lineage 3. Another sample collected as also produced lobed unicellular holdfasts (Fig. 2, u E. foliiformis from Wellington, New Zealand, on the and v). Occasionally, cells had two pyrenoids original host (Lessonia variegata J. Agardh) and type (Fig. 2x) possibly before cell division. The final E. area (South and Adams 1976), did not group with carnea lineage 6 produced multilobed multicellular this monostromatic sample from Australia in the rhizoids (Fig. 2, y and z) and contained samples psbA tree (Fig. S2). Samples from New Zealand from Australia and Great Britain. (4237, 4240, 4772) formed another distinct lineage Often in culture and the field, multiple uprights in the psbA and partial SSU trees (Figs. S2 and S3), were seen originating from one apparent hold- which was not shown in the combined data set fast ⁄ disk. While this is a characteristic of Porphyrostro- because of failure to amplify the rbcL gene. Two mium (see below), in many of our isolates it appears samples identified as Erythrotrichia (E. ligulata and to be a tendency for spores to aggregate (Fig. 2aa) E. pulvinata) were grouped with Porphyrostromium. or stick together on young (Fig. 2bb) or mature Morphological analysis. Observations of isolates in thalli (Fig. 2cc), thus producing multiple shoots unialgal culture showed some subtle differences from a single basal area or the appearance of between some samples from the various lineages of branching. This spore attachment could lead to Erythrotrichia. Lineage 1 was characterized by spores complications in analyzing morphology without that germinated to produce either a digitate (mul- careful observations. tilobed) unicellular holdfast or an elongate rhizoid Porphyrostromium formed a moderately supported (Fig. 2, a and b). This initial cell division also estab- phylogenetic lineage (Fig. 1) and was characterized lished the polarity of the thallus, with the basal cell by a spore germinating as a more or less expanded producing the holdfast, while the upper cell pro- disk from which uprights arise. The early germina- duced the upright reproductive filament. All tion patterns of these isolates mostly consist of unbranched uniseriate filaments had the character- spores that divide anticlinally to the attachment sur- istic oblique cell divisions to produce lenticular cells face producing two- and then four-celled crusts (Fig. 2c) that released monospores. Uniquely, the (Fig. 3b). Continued cell division produced a multi- two or three basal cells of the filament were very cellular crust with bifurcate marginal cells (Fig. 3, b elongate (Fig. 2d), a character not seen in the other and c). These crusts then produced uprights lineages. These elongate basal cells were obconic, (Fig. 3, c and d) that are at first uniseriate, but soon were wider toward the apex, could be up to three became irregularly multiseriate and cylindrical in times as long as wide, and did not produce spores. P. ciliare (Fig. 3d). The early germination pattern No multicellular basal disks were formed in this was similar in P. boryanum (Fig. 3f), which also pro- group. Lineage 2 samples showed similar germina- duced an enlarged crust with bifurcate marginal tion patterns as in lineage 1 (Fig. 2e), but the mul- cells (Fig. 3, f and g) and uprights that were initially tilobed basal holdfast became multicellular in older uniseriate, but then form a monostromatic blade plants (Fig. 2, f and g). Figure 2j shows the general (Fig. 3h). Another isolate within this Porphyrostromi- morphology of E. welwitschii in culture. Again spores um lineage was collected as E. pulvinata. This cul- in culture produced either an elongate rhizoid or a ture, while producing a small flat disk (Fig. 3, i and multilobed disk (Fig. 2i). Distinctly, this species pro- j) and later a pulvinate crust with bifurcate marginal duced branched upright filaments and a multicellu- cells (Fig. 3k), did not produce upright filaments or lar holdfast (Fig. 2j). In older cultures, filaments the characteristic monostromatic upright blades could become irregularly pluriseriate. seen in most field specimens. Lineage 3 was unusual as it contains both a very distinct isolate (E. foliiformis [isolate 4058], see above; Fig. 2, k–m) and isolates with E. carnea mor- DISCUSSION phology (Fig. 2, n–p). This lineage contained sam- Our data on three genes show a mostly supported ples only from the Southern Hemisphere (Australia, phylogeny of the upright Erythropeltidales. The Brazil, Madagascar). These isolates all produced a upright habit evidently arose twice in the order, multicellular holdfast (Fig. 2o) or attach with multi- once in Porphyropsis and again in the samples con- ple rhizoids arising from basal cells as in E. foliifor- cerned in this study. Alternately, the upright habit mis (Fig. 2k). Lineage 4 only contained isolates could be the ancestral character state and lost three from Australia. Spore germination produced either times, in Erythrocladia, Pseudoerythrocladia, and Sahlin- contorted rhizoids (Fig. 2q) or a lobed unicellular gia. One characteristic that distinguishes Porphyropsis disk (Fig. 2, r and s). The first basal cell of the from these other samples is the lack of pyrenoids in upright filament was not elongate (Fig. 2s); this the former (Zuccarello et al. 2010). A sister group 632 GIUSEPPE C. ZUCCARELLO ET AL.

FIG. 2. Isolates of Erythrotrichia in the seven lineages identified in this study (see Fig. 1). Erythrotrichia sp. (lineage 1): (a–c) culture 4398-1, (d) culture 4697. (a, b) First division of spores giving rise to two dissimilar cells, the bottom one developing a multilobed unicellular disk or an elongate rhizoid and the second cell producing the erect filament. (c) Oblique division in monosporangium formation (arrow). (d) Very elongate first cells of the erect filament characteristic of this group. No multicellular basal disks observed in this group. Scale bar is 10 lm in all. Erythrotrichia carnea (lineage 2): (e) culture 4425, (f) culture 4056, (g) culture 4054. (e) Spore germination pattern as in lineage 1 (Fig. 1). In this group, the multilobed basal cell can develop into a multicellular disk in older cultures (f, g). Scale bars: 10 lm in (e) and (g), 15 lm in (f). Erythrotrichia welwitschii: (h) culture 4716, (i) culture 4704, (j) culture 4705. (h) Oblique cell divisions in monosporangial formation (arrows). (i) First division of spore similar to that in the preceding groups, characteristic of all Erythro- trichia isolates. (j) Typical branching of erect filaments and the basal disk that has become multicellular. Scale bars: 10 lm in (h), 15 lm in (i), 20 lm in (j). Erythrotrichia foliiformis morphology from Australia, in lineage 3, culture 4058: (k) the multiple rhizoid-like cells at the base of erect filaments, (l) the contorted blades in older cultures, (m) details of these blades. Scale bars: 15 lm in (k), 100 lm in (l), 10 lm in (m). Erythrotrichia sp. (lineage 3): (n, o) culture 4482, (p) culture 4488. (n, o) The usual germination pattern, in (o) the lobed basal cells are starting to form a multicellular disk. (p) Part of erect filament with plastids dividing before cell division (arrows). Scale bars: 10 lm in (n) and (p), 15 lm in (o). Erythrotrichia sp. (lineage 4): (q, t) culture 4398-2, (r) culture 4393-1, (s) culture 3438. (q, r) The usual pattern of germination although contorted rhizoids are more frequent than lobed disks. (s) Enlarged basal disks, the first cell of the upright filament is short, in the upper filament, the plastids have divided but the cell wall is not formed (arrow). (t) Sporulating filament with oblique divisions (arrows) and discharge papillae. Scale bars: 10 lm in all. Erythrotrichia sp. (lineage 5): (u) culture 4247, (v) culture 4475, (w) culture 4396, (x) culture 4295. (u, v) Germinating spores developing lobed unicell disk, in (v) lobed disk on two sporelings at left and expanding adhesive rhizoid tip at right. (w) Long rhizoid with adhesive expanding tip. (x) The multilobed plastid with one central pyrenoid in mature filaments. Cells with two opposing pyrenoids (arrows). Scale bars: 10 lm in all. Erythrotrichia sp. (lineage 6): (y) culture 4392, (z) culture 4473. (y) Sporulating filament, with oblique partitions (arrows) and a young filament with a lobed unicellular base. (z) Mature filament with a multicellular base. Scale bars: 10 lm in both. ‘‘Sticky spores’’ of Erythrotrichia: (aa) culture 4398-1 (lineage 4), (bb) culture 4645 (lineage 2), (cc) culture 4392 (lineage 6). These figures show what often is seen in cultures of Erythrotrichia. (aa) Spores have settled together giving the impression they have developed as multiple erect filaments from a single basal cell. (bb) Spores are attached to erect filaments that later can give the impression the filament is branched. Also a rhizoid is shown with an expanded tip (arrow) attached to glass. The same rhizoid also is developing a lateral near the erect filament (arrowhead). (cc) Spores germinating on mature filament. Scale bars: 10 lm in (aa) and (cc), 20 lm in (bb). ERYTHROPELTIDALES PHYLOGENY 633

FIG. 3. Isolates of Porphyrostromium investigated in this study. Porphyrostromium ciliare: (a) culture 4654, (b–d) culture 4708. (a) Erect thallus showing cells with a stellate plastid and central pyrenoid. (b, c) Spores divide giving rise to two identical prostrate cells, which in turn divide into two, forming small regular multicellular disks with bifurcate marginal cells. (d) Erect uniseriate filaments grow from these multicellular disks. These filaments quickly become multiseriate, irregular to cylindrical. Scale bars: 10 lm in (a, b), 20 lm in (c), 30 lm in (d). Porphyrostromium boryanum: (e) culture 4703, (f–h) culture 4710. (e) Part of a medium-sized blade with regular cells, and two free spores. (f) Young disks with the unfocused region in the center corresponding to developing erect filaments. Bifircate marginal cells evident. (g) Marginal cells and erect cells in disk center. (h) Older disk with uniseriate filaments and some young multiseriate blades. Scale bars: 15 lm in (e), 30 lm in (f), 10 lm in (g), 70 lm in (h). Porphyrostromium pulvinatum (culture 4758): (i) spore germination on glass, one, two, and four cells. Second division always at right angle to first. (j) Eight-cell stage on Codium. (k) Mature cushions. Note that marginal cells are bifurcate like those of basal disks of other Porphyrostromium and Sahlingia. Scale bars: 10 lm in all. to the upright samples is the crustose genus Sahlingia, mium is unequivocal. We therefore propose two new which has pyrenoids and produces crusts with bifur- combinations. cate marginal cells (Zuccarello et al. 2010), as is Porphyrostromium ligulatum (Womersley) J. A. West seen in isolates of Porphyrostromium. Although our et Zuccarello comb. nov. data involve the collection of many isolates from Basionym: Erythrotrichia ligulata Womersley. 1994. around the world and comprehensive molecular The Marine Benthic Flora of South Australia. Part IIIA. analyses, it does not simplify the taxonomy of the pp. 28–29, fig. 2, E–H). Erythropeltidales, which is at present based on mor- The holotype was collected on Heterozostera in phological characteristics. It is well known that drift at the harbor in Warrnambool, Victoria, Austra- genetic variation and phylogenetic history is not lia. It is also known from other hosts and locations always reflected in morphological change. This is in Tasmania and South Australia (Womersley 1994). especially evident in ‘‘simple’’ red algae (e.g., Ban- Womersley’s illustrations clearly show a large basal gia Nelson et al. 2005, Acrochaetium West et al. 2008, disk with upright monostromatic blades, having cells Bostrychia Zuccarello and West 2006). Therefore, the with a single chloroplast and pyrenoid and mono- taxonomy of the upright Erythropeltidales could be sporangia typical of most Erythropeltidales. We col- revised based upon our molecular results, and we lected similar specimens on Heterozostera at the partially do this here. harbor in Warrnambool on 28 November 2007 and It appears that Porphyrostromium can be morpholog- dried them for molecular analyses. Unfortunately, ically and phylogenetically recognized as a distinct the live specimens did not survive to grow in genus, even though its phylogenetic position within culture. the other upright isolates is not fully resolved. The Porphyrostromium pulvinatum (N. L. Gardner) J. A. development of a multicellular crust (Kikuchi and West et Zuccarello comb. nov. Shin 2011) and multiple uprights arising from this Basionym: Erythrotrichia pulvinata N. L. Gardner. crust appear to be a major characteristic of this genus 1927. New Rhodophyceae from the Pacific coast of and distinguish it from Erythrotrichia. The presence of North America. II. University of California Publications bifurcate marginal cells is a feature that could lead to in Botany 13: 238, pl. 24, figs. 1–3. its misidentification as Sahlingia in its early stages E. pulvinata is recorded from Pacific North Amer- (i.e., before uprights form). Although the position of ica as an epiphyte on C. fragile utricles (Dawson E. pulvinata was only based on one gene (psbA; 1953, McBride 1972, Abbott and Hollenberg 1976, Fig. S2), its position with other species of Porphyrostro- Scagel et al. 1993, Hansen 1997). Our cultured 634 GIUSEPPE C. ZUCCARELLO ET AL. specimens have not formed uniseriate and multiseri- 6), the holdfast becomes multicellular and can ate blades. This characteristic was also noted in cul- resemble a Porphyrostromium disk. This issue shows ture specimens by McBride (1972), and Gardner the importance of studying the ontogeny to distin- (1927) stated that some field specimens lacked guish these two groups. blades. Nonetheless, the clearly defined molecular One lineage within Erythrotrichia, E. welwitschii,isa evidence and the bifurcate disk marginal cells support recognized species. This species is distinct with a the placement of this species in the genus Porphyrostro- branched thallus (Brodie and Irvine 2003) (Fig. 2j). mium. A key to the four species of Porphyrostromium In the field, this species grows on Ralfsia sp., which investigated in this study is presented in Table 1. in turn grows on mollusk shells (e.g., Patella sp.), Before continuing, we would like to briefly men- but also on rocks. This unusual habitat leads them tion the placement of other genera and samples in the field to have an elongate germination rhizoid (e.g., samples from GenBank) within the upright that penetrates between the filaments of Ralfsia sp. lineage. As data with all three genes were not avail- (Brodie and Irvine 2003). able with these samples, they were not used in the Except for E. welwitschii, the lineages of Erythrotri- three-gene data set. Our analysis suggests that chia are more problematic because defining features S. naiadum (C. L. Anderson) Hollenb., Pyrophyllon are not evident. All the lineages have a bipolar ger- W. A. Nelson, and Chlidophyllum kaspar (W. A. Nel- mination pattern and uniseriate erect filaments son et N. M. Adams) W. A. Nelson are distinct mem- (except in lineage 3, see below) and lack branching. bers of the Erythropeltidales (Rintoul et al. 1999, Lineage 1 is the most distinct lineage, having 1–3 Nelson et al. 2003, West and Zuccarello 2009). C. elongate basal cells (Fig. 2d), while all the other kaspar and Pyrophyllon form a well-supported mono- lineages have basal cells approximately equal in phyletic group (Figs. S1 and S3) as does S. naiadum length to other cells. This character is most evident (Fig. S2); however, the sister relationships of these in mature attached filaments in our cultures. There groups to other lineages are still not clear. Our data are also differences in holdfast morphology. While also show that there is a distinct lineage of E. car- holdfasts can be rhizoidal or digitate single cells in nea–like samples from New Zealand [isolates 4237, some cultures, these, with age, become multicellu- 4240, and 4772 and GenBank samples (Nelson et al. lar. Multicellular holdfasts are not seen in lineage 1, 2003)] (Figs. S2 and S3). but other lineages usually have multicellular hold- It is evident that Erythrotrichia sensu lato consists fasts depending on plant age. of at least seven well-supported lineages. The devel- We do have an unusual outcome that within line- opment of the isolates in this group is distinct from age 3, in which mainly uniseriate isolates with digi- Porphyrostromium. The germination pattern of Erythro- tate holdfasts are seen, one isolate is identified as trichia is characterized by spores that germinate in a E. foliiformis (isolate 4058) with monostromatic blades bipolar manner to produce two cells; the lower cell and rhizoidal holdfast. It is possible that uniseriate or produces an elongate rhizoid or digitate unicellular multiseriate blades are an easily gained character, holdfast, and the upper cell produces the upright which calls into question the monophyly of species of filament. This step mostly leads to a single upright Erythrotrichia that are described based on filaments per spore. This pattern is obscured by the attach- that are 2 to 8 cells across [e.g., Erythrotrichia bertholdii ment of spores to parent plants or the aggregation Batters, Erythrotrichia investiens (Zanardini) Bornet]. of spores possibly due to their gliding together This isolate (4058) is not genetically similar to E. folii- (Pickett-Heaps et al. 2001). This complication may formis from the type area (New Zealand, F582) lead to the misidentification of multiple uprights (Fig. S2). We suggest that records of E. foliiformis emanating from a single location as a species of from New South Wales be reinvestigated. Porphyrostromium in which multiple uprights can Can we attribute names to these phylogenetic arise from a single crust. Also, in older plants, in lineages, especially the commonly used name E. car- some isolates of Erythrotrichia (e.g., lineage 2, lineage nea? There are 27 current names of Erythrotrichia in Algaebase (Guiry and Guiry 2010; http://www.algae base.org; searched on 1 June 2010); while a few of these are not the entities studied as they are not TABLE 1. Key to the species of Porphyrostromium investigated in this study. uniseriate (e.g., E. bertholdii, but see above), the attachment of the name E. carnea (or any other 1. Mature thalli cylindrical <120 lm wide (to 5–8 cells across) name) to any one lineage is problematic. The prob- to 1–2 mm long ...... Porphyrostromium ciliare lem is further compounded as the species descrip- 1. Mature thalli monostromatic, flat, 0.2–4 mm wide ...... 2 tions in the literature are minimal and sometimes 2. Mature thalli flat, narrow, up to 250 lm wide to 1–2 cm contradictory, and key features (e.g., holdfast mor- long (Korea and Japan) ...... Porphyrostromium japonicum 2. Mature thalli flat, wider than 200 lm ...... 3 phology, basal cell dimensions) are missing. Our 3. Mature thalli flat, narrow, to 200–600 lm wide, <4–8 mm observations on isolates in culture do not clearly dis- long (Australia) ...... Porphyrostromium ligulatum tinguish the six lineages from each other. More 3. Mature thalli flat, wide, up to 4 mm wide, to 1–3 cm importantly, multiple lineages are found at the same long ...... Porphyrostromium boryanum collection sites. Therefore, location (type locality) ERYTHROPELTIDALES PHYLOGENY 635 does not help. We often isolated different Erythrotri- Reasons: We designated an epitype for the follow- chia lineages from the same substrate on the same ing reasons: multiple molecular lineages, which day in the same location, and it is clear that most would warrant a taxonomic rank, all appear mor- lineages show a wide distribution (i.e., no biogeo- phologically identical to E. carnea, so a lineage can- graphic patterns are seen). S. alsidii is another case not be designated from dried specimens. These in which small red algae seem to be truly ubiquitous E. carnea lineages can be found on the same sub- (Zuccarello et al. 2008). We believe that multiple strate in the same locality. There is a very low proba- local collections from most areas will reveal many of bility of gathering DNA from the type specimen these lineages in each area. (due to its age), and as multiple lineages are likely So how do we assign names, and should we, espe- on the basiphyte, a specific evolutionary lineage cially assign the most common name, E. carnea,to may also be difficult to designate. Therefore, for our lineages? Shall we lump or split? This problem future taxonomic stability, we designate E. carnea has been tackled before with simple algae (e.g., Chla- isolate 4056 as the epitype, and further taxonomic mydomonas; Pro¨schold et al. 2001) and is increasing revision will be forthcoming. in prevalence as genetic data continue to unravel This leaves several lineages (3–6) in need of a high levels of diversity in morphological look-alikes new generic name, but with no diagnostic charac- (cryptic species; Boo et al. 2010). We could select a ters. We are reluctant to name new genera based lineage that contains samples close to the type local- only on limited gene sequence data and no extra ity (Glamorgan, Wales, Great Britain) as E. carnea, morphological analysis (e.g., ultrastructural). and this would be lineage 2 (isolate 4056 from Gow- Based on their level of variation, we could also er, Wales). However, the lack of any known biogeo- designate seven new genera for the lineages, which graphic pattern of these entities makes this reflect genetic evolution, but not morphological approach also a little problematic (i.e., the lineage evolution. Conversely, we could call them all E. car- newly sampled as the type locality could just as easily nea and characterize it as a clade containing many be another of the lineages with E. carnea morphol- entities that can be described by their phylogenetic ogy). In fact, it is possible that most of the several placement (using three genetic markers) into dis- lineages revealed are truly cosmopolitan, as seen in tinct clades (as a phylocode approach). As genetic S. alsidii (Zuccarello et al. 2008). data become more essential to designate species Maybe we should start again and not try to untan- (e.g., cryptic species, nonreproductive specimens), gle poor descriptions, morphologically similar mate- we believe nomenclatural rules may become more a rial, and type material from which DNA analysis is burden than a help. unlikely to be fruitful (due to age, size, and multiple In conclusion, our culture studies and molecular lineages on identical substrates). If we want to keep analysis support the recognition of Porphyrostromium the monophyly of Erythrotrichia, which should as a distinct genus and the transfer of two species of include E. welwitschii as it is the only lineage we can Erythrotrichia to this genus. The lack of biogeography confidently assign a name to, then the clade contain- in the lineages of E. carnea makes the designation of ing lineages 1 and 2, and E. welwitschii (and isolate a type problematic, but we suggest a sample in line- 4474) could be assigned to Erythrotrichia. We there- age 2 collected close to the type locality be desig- fore designate a specimen from lineage 2 (isolate nated as the epitype. Our data highlight the 4056) collected from Gower, Wales, close to the type problems of species taxonomy in which morphology locality, as the epitype for the genus (below and is not a good designator, and we suggest that future Fig. S4 in the supplementary material) and keep E. taxonomies, especially of morphologically simple welwitschii as an Erythrotrichia, and lineage 1 samples organisms, may have to be based exclusively on phy- of an as yet unnamed species of Erythrotrichia. logenetic results. Type: Erythrotrichia carnea (Dillwyn) J. Agardh (1883). Till algernes systematik. Nya bidrag. (Tredje We thank Juliet Brodie, Norio Kikuchi, Tracy Farr, Erasmo ˚ Macaya, and Silvia Guimara˜es for supplying samples. This afdelningen.). Lunds Universitets Ars-Skrift, Afdelningen research was partially sponsored by the Australian Research for Mathematik och Naturvetenskap 19(2): 15. Council, Australian Biological Resources Study, and the Type location: BM, Glamorgan (Loughor), Wales, Hermon Slade Foundation. This project was also partially Great Britain. September 1805, coll. W. W. Young. supported by the National Science Foundation to H. S. Y. Epitype: Erythrotrichia carnea, isolate J. A. West cul- (EF 08-27023, DEB 09-37975). ture 4056. Port Eynon (5132¢ N, 0412¢ W), Gower, Wales, Great Britain, collected by Juliet Brodie on Abbott, I. A. & Hollenberg, G. J. 1976. Marine Algae of California. Laurencia sp. 209, 5 September 1997. Figure S4 (in Stanford University Press, Stanford, California, 827 pp. the supplementary material). Bass-Becking, L. G. M. 1934. Geobiologie of Inleiding tot de Milieikunde. Deposited: NSW865944, Royal Botanic Gardens, Van Stockum & Zon, The Hague, the Netherlands, 263 pp. Boo, S. M., Kim, H. S., Shin, W., Boo, G. H., Cho, S. M., Jo, B. Mrs. Macquaries Road, Sydney, NSW 2000, Australia. Y., Kim, J.-H., et al. 2010. Complex phylogeographic patterns Live type culture: CCAP 1369 ⁄ 5. Culture Collection in the freshwater alga Synura provide new insights into of Algae and Protozoa. SAMS Research Services Ltd. ubiquity vs. endemism in microbial eukaryotes. Mol. Ecol. Dunbeg, Argyll, PA37 1QA, UK. 19:4328–38. 636 GIUSEPPE C. ZUCCARELLO ET AL.

Brodie, J. A. & Irvine, L. M. 2003. Volume 1 Rhodophyta, Part 3B for obligate epiphytic species previously placed in Porphyra, Bangiophycidae. Intercept Limited, Andover, UK, 167 pp. and a discussion of the orders Erythropeltidales and Bangiales. Broom, J. E., Jones, W. A., Hill, D. F., Knight, G. A. & Nelson, W. A. Phycologia 42:308–15. 1999. Species recognition in New Zealand Porphyra using 18S Nelson, W. A., Farr, T. J. & Broom, J. E. S. 2005. Dione and Minerva, rDNA sequencing. J. Appl. Phycol. 11:421–8. two new genera from New Zealand circumscribed for basal Dawson, E. Y. 1953. Marine red algae of Pacific Mexico. Part 1. taxa in the Bangiales (Rhodophyta). Phycologia 44:139– Bangiales to Corallinaceae subf. Corallinoidae. Allan Hancock 45. Pac. Exp. 17:1–239. Pickett-Heaps, J. D., West, J. A., Wilson, S. M. & McBride, D. L. Felsenstein, J. 1985. Confidence intervals on phylogenies: an 2001. Time-lapse videomicroscopy of cell (spore) movement approach using the bootstrap. Evolution 39:783–91. in red algae. Eur. J. Phycol. 36:9–22. Fenchel, T. & Finlay, B. J. 2004. The ubiquity of small species: Pro¨schold, T., Marin, B., Schlo¨sser, U. G. & Melkonian, M. 2001. patterns of local and global diversity. Bioscience 54:777–84. Molecular phylogeny and taxonomic revision of Chlamydo- Finlay, B. J., Esteban, G. F. & Fenchel, T. 2004. Protist diversity is monas (Chlorophyta). I. Emendation of Chlamydomonas Eh- different? Protist 155:15–22. renberg and Chloromonas Gobi, and description of Gardner, N. L. 1927. New Rhodophyceae from the Pacific coast of Oogamochlamys gen. nov. and Lobochlamys gen. nov. Protist North America. VI. Univ. Calif. Publ. Bot. 14:99–138. 152:265–300. Guiry, M. D. & Guiry, G. M. 2010. AlgaeBase. World-wide electronic Rambaut, A. 1996. Se-Al: Sequence Alignment Editor. Available at: publication, National University of Ireland, Galway. Available http://evolve.zoo.ox.ac.uk/. at: http://www.algaebase.org (last accessed 10 October 2010). Rintoul, T. L., Sheath, R. G. & Vis, M. L. 1999. Systematics and Hansen, G. I. 1997. A revised checklist and preliminary assessment biogeography of the Compsopogonales (Rhodophyta) with of the macrobenthic marine algae and seagrasses of Oregon. emphasis on the freshwater families in North America. Phyco- In Kaye, T. N., Liston, A., Love, R. M., Luoma, D. L., Meinke, logia 38:517–27. R. J. & Wilson, M. V. [Eds.] Conservation and Management of Saunders, G. W. & Kraft, G. T. 1994. Small-subunit rRNA Native Flora and Fungi. Corvallis Native Plant Society of Oregon, gene sequences from representatives of selected families of Corvallis, Oregon, pp. 175–200. the Gigartinales and Rhodymeniales (Rhodophyta). 1. Hawkes, M. W. 1988. Evidence of sexual reproduction in Smithora Evidence for the Plocamiales ord. nov. Can. J. Bot. 72:1250– naiadum (Erythropeltidales, Rhodophyta) and its evolutionary 63. significance. Br. Phycol. J. 23:327–36. Scagel, R. F., Gabrielson, P. W., Garbary, D. J., Golden, L., Hawkes, Heerebout, G. R. 1968. Studies on the Erythropeltidaceae (Rhod- M. W., Lindstrom, S. C., Oliveira, J. C. & Widdowson, T. B. ophyceae, Bangiophycidae). Blumea 16:139–57. 1993. A Synopsis of the Benthic Marine Algae of British Columbia, van den Hoek, C., Mann, D. G. & Jahns, H. M. 1995. Algae. An Southeast Alaska, Washington and Oregon. Phycological Contri- Introduction to Phycology. Cambridge University Press, Cam- bution No. 3. Department of Botany, University of British bridge, UK, 623 pp. Columbia, vol. 3, pp. 1–532. Huelsenbeck, J. P. & Ronquist, F. 2001. MRBAYES: Bayesian Scott, J. L., Orlova, E. & West, J. A. 2010. Ultrastructural observa- inference of phylogenetic trees. Bioinformatics 17:754–5. tions of vegetative cells of two new genera in the Erythro- Karsten, U., West, J. A., Zuccarello, G. C., Engbrodt, R., Yokoyama, peltidales (Compsopogonophyceae, Rhodophyta): A., Hara, Y. & Brodie, J. 2003. Low molecular weight carbo- Pseudoerythrocladia and Madagascaria. Algae 25:11–5. hydrates of the Bangiophyceae (Rhodophyta). Part II. J. Phycol. Silva, P. C., Basson, P. W. & Moe, R. L. 1996. Catalogue of the 39:584–9. benthic marine algae of the Indian Ocean. Univ. Calif. Publ. Karsten, U., West, J. A., Zuccarello, G. C., Nixdorf, O., Barrow, K. D. Bot. 79:1–1256. & King, R. J. 1999. Low molecular weight carbohydrate pat- South, G. R. & Adams, N. M. 1976. Erythrotrichia foliiformis sp. nov. terns in the Bangiophyceae (Rhodophyta). J. Phycol. 35:967– (Rhodophyta, Erythropeltidaceae) from New Zealand. J. R. 76. Soc. N. Z. 6:399–405. Kikuchi, N. & Shin, J.-A. 2011. Porphyrostromium japonicum (Tokida) Stamatakis, A. 2006. RAxML-VI-HPC: maximum likelihood-based Kikuchi comb. nov. (Erythropeltidales, Rhodophyta) from Ja- phylogenetic analyses with thousands of taxa and mixed pan. Phycologia 50:122–31. models. Bioinformatics 22:2688–90. Kornmann, P. 1984. Erythrotrichopeltis, a new genus of the Erythro- Swofford, D. L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony peltidaceae (Bangiophyceae, Rhodophyta). Helgol. Meeresun- (*and Other Methods). Sinauer Associates, Sunderland, Massa- ters. 38:207–24. chusetts. Kornmann, P. 1987. The life cycle of Porphyrostromium obscurum West, J. A. 2005. Long term macroalgal culture maintenance. In (Bangiophyceae, Rhodophyta). Helgol. Meeresunters. 41:127–37. Andersen, R. A. [Ed.] Algal Culturing Techniques. Academic Kornmann, P. 1989. Sahlingia nov. gen. based on Erythrocladia sub- Press, New York, pp. 157–63. integra (Erythropeltidales, Rhodophyta). Br. Phycol. J. 24:223–8. West, J. A., Scott, J. L., West, K. A., Karsten, U., Clayden, S. L. & Magne, F. 1960. Le Rhodochaete parvula Thuret (Bangioideé) et sa Saunders, G. W. 2008. Rhodachlya madagascarensis gen. et sp reproduction sexue. Cah. Biol. Mar. 1:407–20. nov.: a distinct acrochaetioid represents a new order Magne, F. 1990. Reproduction sexueé chez Erythrotrichia carnea and family (Rhodachlyales ord. nov., Rhodachlyaceae fam. (Rhodophyceae, Erythropeltidales). Cryptogam. Algol. 11:157– nov.) of the Florideophyceae (Rhodophyta). Phycologia 70. 47:203–12. McBride, D. L. 1972. Studies on some British Columbian repre- West, J. A. & Zuccarello, G. C. 1999. Biogeography of sexual and sentatives of the Erythropeltidaceae (Rhodophyceae, Bang- asexual populations in Bostrychia moritziana (Rhodomelaceae, iophycidae). PhD thesis, University of British Columbia, 108 Rhodophyta). Phycol. Res. 47:115–23. pp. West, J. A. & Zuccarello, G. C. 2009. Taxonomic deflation: molec- Mu¨ller, K. M., Oliveira, M. C., Sheath, R. G. & Bhattacharya, D. ular evidence places Membranella nitens in taxonomic synonymy 2001. Ribosomal DNA phylogeny of the Bangiophycidae with Smithora naiadum (Erythropeltidales, Compsopogono- (Rhodophyta) and the origin of secondary plastids. Am. J. Bot. phyceae, Rhodophyta). Phycologia 48:543–6. 88:1390–400. West, J. A., Zuccarello, G. C., Scott, J. L., West, K. A. & de Goer, S. L. Murray, S. W., Dixon, P. S. & Scott, J. L. 1972. The life history of 2007a. Pulvinus veneticus gen. et sp. nov. (Compsopogonales, Porphyropsis coccinea var. dawsonii in culture. Br. Phycol. J. 7:323– Rhodophyta) from Vanuatu. Phycologia 46:237–46. 33. West, J. A., Zuccarello, G. C., Scott, J. L., West, K. A. & Loiseaux Nelson, W. A. 1993. Epiphytic species of Porphyra (Bangiales, Rho- de Goe¨r, S. 2007b. Corrigendum. Correction to paper by dophyta) from New Zealand. Bot. Mar. 36:525–34. West et al., Phycologia 46:237–46 (2007). Pulvinaster venetus Nelson, W. A., Broom, J. E. & Farr, T. J. 2003. Pyrophyllon and J.A. West, G.C. Zuccarello & J.L. Scott gen. et sp. nov. Phyco- Chlidophyllon (Erythropeltidales, Rhodophyta): two new genera logia 46:478. ERYTHROPELTIDALES PHYLOGENY 637 de Wit, R. & Bouvier, T. 2006. ‘Everything is everywhere, but, the environment selects’; what did Baas Becking and Beijerinck Supplementary Material really say? Environ. Microbiol. 8:755–8. Womersley, H. B. S. 1994. The Marine Benthic Flora of Southern Aus- The following supplementary material is avail- tralia. Part IIIA. State Herbarium of South Australia, South able for this article: Australia, 508 pp. Wynne, M. J. 1986. Porphyrostromium Trevisan (1848) vs Erythrotri- Figure S1. Phylogeny of the Erythropeltidales chopeltis Kornmann (1984) (Rhodophyta). Taxon 35:328–9. based on rbcL gene data set. Yokoyama, A., Scott, J. L., Zuccarello, G. C., Kajikawa, M., Hara, Y. & West, J. A. 2009. Corynoplastis japonica gen. nov. and Dixoniell- ales ord. nov. (Rhodellophyceae, Rhodophyta) based on Figure S2. Phylogeny of the Erythropeltidales morphological and molecular evidence. Phycol. Res. 57:278–89. based on psbA gene data set. Yoon, H. S., Hackett, J. D., Pinto, G. & Bhattacharya, D. 2002. The single, ancient origin of chromist plastids. Proc. Natl. Acad. Sci. Figure S3. Phylogeny of the Erythropeltidales U. S. A. 99:15507–12. based on partial SSU gene data set. Yoon, H. S., Mu¨ller, K. M., Sheath, R. G., Ott, F. D. & Bhattacharya, D. 2006. Defining the major lineages of red algae (Rhodo- Figure S4. Erythrotrichia carnea (isolate 4056), phyta). J. Phycol. 42:482–92. designated the epitype. Scale bar in all figures is Zuccarello, G. C., Kikuchi, N. & West, J. A. 2010. Molecular phylogeny of the crustose Erythropeltidales (Compsopogonophy- 10 lm. ceae, Rhodophyta): new genera Pseudoerythrocladia and Madagascaria and the evolution of the upright habit. J. Phycol. Table S1. Samples of the Erythropeltidales 46:363–73. used in molecular analysis. Zuccarello, G. C. & Lokhorst, G. M. 2005. Molecular phylogeny of the genus Tribonema (Xanthophyceae) using rbcL gene This material is available as part of the online sequence data: monophyly of morphologically simple algal article. species. Phycologia 44:384–92. Zuccarello, G. C. & West, J. A. 2006. Molecular phylogeny of the Please note: Wiley-Blackwell are not responsi- subfamily Bostrychioideae (Ceramiales, Rhodophyta): subsuming Stictosiphonia and highlighting polyphyly in species of ble for the content or functionality of any sup- Bostrychia. Phycologia 45:24–36. porting materials supplied by the authors. Any Zuccarello, G., West, J., Bitans, A. & Kraft, G. 2000. Molecular queries (other than missing material) should be phylogeny of Rhodochaete parvula (Bangiophycidae, Rhodo- directed to the corresponding author for the phyta). Phycologia 39:75–81. Zuccarello, G. C., West, J. A. & Kikuchi, N. 2008. Phylogenetic article. relationships within the Stylonematales (Stylonematophyceae, Rhodophyta): biogeographic patterns do not apply to Stylo- nema alsidii. J. Phycol. 44:384–93.