The Coralline Genera Sporolithon and Heydrichia (Sporolithales, Rhodophyta) Clarified by Sequencing Type Material of Their Generitypes and Other Species1

Home , RedToL, Rhodogorgonales, Spongites yendoi, Sporolithaceae

THE CORALLINE GENERA SPOROLITHON AND HEYDRICHIA (SPOROLITHALES, RHODOPHYTA) CLARIFIED BY SEQUENCING TYPE MATERIAL OF THEIR GENERITYPES AND OTHER SPECIES1

Joseph L. Richards 2 Department of Biology, University of Louisiana at Lafayette, Lafayette, Louisiana 70504-3602, USA Thomas Sauvage Smithsonian Marine Station at Fort Pierce, Fort Pierce, Florida 34949-3140, USA Department of Biology, University of Louisiana at Lafayette, Lafayette, Louisiana 70504-3602, USA William E. Schmidt, Suzanne Fredericq Department of Biology, University of Louisiana at Lafayette, Lafayette, Louisiana 70504-3602, USA Jeffery R. Hughey Division of Mathematics, Science, and Engineering, Hartnell College, Salinas, California 93901, USA and Paul W. Gabrielson Biology Department and Herbarium, University of North Carolina at Chapel Hill, Coker Hall CB 3280, Chapel Hill, North Carolina 27599-3280, USA

Interspecific systematics in the red algal order reported to be H. woelkerlingii from New Zealand is Sporolithales remains problematic. To re-evaluate likely an undescribed species. Type specimens of all its species, DNA analyses were performed on other Sporolithon and Heydrichia species need to be historical type material and recently collected sequenced to confirm that they are distinct species; specimens assigned to the two genera Sporolithon morpho-anatomical studies have proved inadequate and Heydrichia. Partial rbcL sequences from the for this task. lectotype specimens of Sporolithon ptychoides (the Key index words: COI; Egypt; Gulf of Suez; LSU; generitype species) and Sporolithon molle, both from new species; phylogenetics; psbA; rbcL; Red Sea; El Tor, Egypt, are exact matches to field-collected UPA topotype specimens. Sporolithon crassum and Sporolithon erythraeum also have the same type Abbreviations: bp, base pairs; BS, bootstrap value; locality; material of the former appears to no longer ML, maximum likelihood; NWGMx, Northwest Gulf exist, and we were unable to PCR amplify DNA of Mexico; SEGMx, Southeast Gulf of Mexico from the latter. A new species, Sporolithon eltorensis, is described from the same type locality. We have not found any morpho-anatomical characters that To determine if any genus is monophyletic and distinguish these three species. No sequenced which species belong in the genus, one must begin specimens reported as S. ptychoides from other parts by characterizing the generitype species. Histori- of the world represent this species, and likely cally, this characterization was done for coralline reports of S. ptychoides and S. molle based on algae by morpho-anatomical comparisons using morpho-anatomy are incorrect. A partial rbcL gross morphology at first, but soon thereafter aug- sequence from the holotype of Sporolithon dimotum mented by anatomical characters through decalcifi- indicates it is not a synonym of S. ptychoides, and cation, sectioning and light microscopy, and finally data from the holotype of S. episporum confirm its in the late 20th century aided by scanning electron specific recognition. DNA sequences from topotype microscopy (Johansen 1981, Woelkerling 1988). material of Heydrichia woelkerlingii, the generitype Recently, this approach has been further augmented species, and isotype material of Heydrichia cerasina by DNA sequence data that has, thus far, supported confirm that these are distinct species; the taxon the monophyly of only one genus, Bossiella (Hind et al. 2014a, 2015), but revealed that many others 1Received 6 March 2017. Accepted 15 June 2017. First Published were polyphyletic, e.g., Calliarthron (Gabrielson et al. Online 6 July 2017. Published Online 7 August 2017, Wiley Online 2011), Clathromorphum (Adey et al. 2015), Corallina Library (wileyonlinelibrary.com). 2Author for correspondence: e-mail [email protected]. (Hind and Saunders 2013), Hydrolithon, Porolithon, Editorial Responsibility: M. Vis (Associate Editor) and Spongites (Rӧsler et al. 2016). Furthermore, the

1044 THE CORALLINE GENERA SPOROLITHON AND HEYDRICHIA 1045 generitype species of three additional genera, Pach- had been considered a later heterotypic synonym of yarthron, Serraticardia and Yamadaia were all shown S. ptychoides since the early 20th century, was more to belong in Corallina, and these genera were placed similar to S. erythraeum, leaving only two equivocal in synonymy by Hind et al. (2014b), Hind and characters to distinguish the three species: (i) size Saunders (2013) and Martone et al. (2012), respec- of tetrasporangia (partially overlapping in length tively. Many other coralline genera have yet to be and width), and (ii) whether or not old, buried assessed using this methodology. terasporangia were sometimes infilled (Verheij DNA sequencing of the type specimen of generi- 1993, table 2). To date, no original material of type species has been critical to correctly applying S. crassum has been located in other herbaria. Thus, these names. For example, Bossiella plumosa, the at present, all four species are recognized, and generitype species and B. frondifera both share the S. erythraeum, S. molle, and S. ptychoides are reported same type locality, and sequencing recently collected to be widely distributed based on morpho-anatomy topotype material did not allow names to be correctly (Guiry and Guiry 2017, accessed January 20, 2017). applied until the holotype specimen of each was The generitype species of Heydrichia, H. woelker- sequenced (Hind et al. 2015). In the case of the lingii, is based on specimens collected from Oudek- generitype species of Lithophyllum, L. incrustans, that raal, Western Cape Province, South Africa name had been misapplied to specimens for 60 years (Townsend et al. 1994). Sequencing of the holo- (Hernandez-Kantun et al. 2015). Western European type, isotype, and other specimens examined in the field-collected specimens with rbcL sequences identi- protologue is likely not possible due to formalin cal to the lectotype specimen were subtidal rhodo- and glutaraldehyde preservation. The first published liths, or less commonly, subtidal, epilithic, epizoic, or SSU sequence of H. woelkerlingii was from a speci- epiphytic crusts, not a common, epilithic intertidal men collected from Betty’s Bay near the type locality crust as widely reported (Cabioch and Boudouresque (Bailey and Chapman 1998). Subsequently, Farr 1992, Chamberlain and Irvine 1994). et al. (2009) provided psbA sequences of specimens Four currently recognized Sporolithon species, identified as H. woelkerlingii from New Zealand. S. ptychoides, the generitype, S. erythraeum, S. crassum, More recently, Mateo-Cid et al. (2014) published a and S. molle, all have their type locality at El Tor, psbA sequence of a specimen collected at the type Egypt in the Gulf of Suez, Red Sea, and all have locality in South Africa. Bahia et al. (2015a) showed intermingled taxonomic histories. Sporolithon ery- that the psbA sequence of a specimen identified as thraeum was originally described as Lithothamnion ery- H. woelkerlingii from New Zealand did not form a thraeum (Rothpletz 1893) based on material found clade with the psbA sequence of the specimen from on the beach. In the protologue of S. ptychoides, two South Africa. forms were described, S. ptychoides f. dura (Heydrich Herein, we apply DNA sequencing methodology 1897a, pl. III, figs. 20–23) and S. ptychoides f. mollis to the generitype species, Sporolithon ptychoides and (Heydrich 1897a, pl. III, figs. 15–19). Heydrich Heydrichia woelkerlingii. For the former, we obtained (1897b) subsequently split the S. ptychoides f. mollis an rbcL sequence from the lectotype specimen; for material and recognized two species, S. molle (Hey- the latter, an rbcL sequence from topotype material. drich 1897a, pl. III, figs. 16, 18, 19) and S. crassum We also sequenced the lectotype of S. molle and (Heydrich 1897a, pl. III, fig. 15), leaving S. pty- the holotypes of Sporolithon dimotum (type local- choides, based on S. ptychoides f. dura as the generi- ity: Lemon Bay, near Guanica, Puerto Rico), cur- type species. Woelkerling and Townsend (in rently considered a synonym of S. ptychoides, and Woelkerling 1988: 207) noted that Heydrich’s origi- of S. episporum (type locality: Point Toro, near nal material of S. ptychoides, S. molle, and S. crassum Colon, Panama). Based on the results of the DNA was destroyed, but that syntype material of the first sequence analyses and SEM observations of mor- two was present in the Foslie herbarium (TRH). pho-anatomical characters, we assess the identity of They designated this syntype material of S. ptychoides species of Sporolithon and Heydrichia and describe a in TRH as the lectotype specimen. Much earlier, new coralline alga, S. eltorensis sp. nov. Foslie (1900;) (as Archaeoliththamnium erythraeum) had considered S. ptychoides to be a later, heterotypic synonym of S. erythraeum, and this was confirmed MATERIAL AND METHODS by Woelkerling and Townsend (in Woelkerling Specimen collection. Collections of Sporolithon spp. specimens 1988: 207) based on a morpho-anatomical compar- were made in the shallow subtidal zone (1–1.5 m deep) near ison of the two type specimens, both in TRH. How- El Tor, Egypt as well as Dahab, Egypt, along fringing reefs ever, Verheij (1993) examined the arrangement of and reef flats during a field expedition focused on assessing tetrasporangia in the type specimens of both S. pty- algal biodiversity in the northern Red Sea (Table 1). Speci- choides and S. molle and, based on the presence or mens of non-geniculate coralline algae were collected by hand while snorkeling and preserved in silica gel; vouchers absence of a distinct layer of elongated cells among were deposited at NCU or LAF (herbarium codes follow the tetrasporangia, determined that they were dis- Thiers 2017). Specimens identified as S. episporum and Sporoli- tinct species, in addition to S. erythraeum. He also thon sp. were collected using SCUBA from the subtidal zone noted that the lectotype specimen of S. molle, which in the Caribbean (3 m) and Pacific Panama (6–7 m), 1046

TABLE 1. List of taxa with collection data and GenBank numbers for newly generated sequences. Sequences included in the concatenated analyses shown in italics.

GenBank accession no. ID no./reason for inclu- Taxa sion Locality Collector/date habitat LSU COI psbA rbcL UPA Antithamnion sp. LAF 4355 phylogeny Ewing Bank, NWGMx S. Fredericq, J. Richards, KY980440 KY994115 – KY994130 – 28°06.0660 N; 91°02.1460 W 29.viii.2011 (collected from site), 25.ii.2012 (collected from microcosm), epilithic, 58–91 m depth Heydrichia cerasina NCU 617165 (UWC l’Agulhas, Cape Agulhas, G. W. Maneveldt 16.vi.2010, on KY980439 KY994114 MF034551 KY994128 – 10/144) Western Cape Province, pebbles in sand, low intertidal Isotype, phylogeny, South Africa distribution 34°4906.09″ S; 20°0107.83″ E Mesophyllum NCU 590286 Wembury, South Devon, J. Brodie, 22.vii.2009, epilithic in ––MF034552 KY994129 – lichenoides Phylogeny England tidepool 50°190 N; 4°040 59.88″ W Sporolithon dimotum NY 900043 (Howe Lemon Bay, near Guanica, M. A. Howe, 27.vi.1903 –––KY994131 –

2667) Puerto Rico AL. ET RICHARDS L. JOSEPH Holotype of Archaeolithothamnium dimotum Sporolithon eltorensis NCU 606659 El Tor, Egypt, Gulf of Suez T. Sauvage, W. E. Schmidt, D. KY980433 – MF034543 –– (LAF 5850) 28°14.0800 N; 33°36.1740 E Gabriel, 8.v.2012, 1–2 m depth Holotype, phylogeny, distribution Sporolithon eltorensis LAF 5767 Dahab, Egypt, Gulf of Aqaba T. Sauvage, W. E. Schmidt, D. KY980434 KY994110 MF034544 – KY980428 (NCU 649164) 28°28.6540 N; 34°30.6980 E Gabriel, 5.v.2012, 1–2 m depth Isotype, phylogeny, distribution Sporolithon NY 900041 (Howe Point Toro, near Colon, M. A. Howe, 7.i.1910, low water –––KY994125 – episporum 6832) Panama, Caribbean Sea mark to several meters depth Holotype of Archaeolithothamnion episporum Sporolithon NCU 598843 Tervi Bight, Bocas del Toro, K. Miklasz, 13.vii.2008, epilithic, – KY994113 MF034547 KY994124 – episporum (PHYKOS 5467) Panama, Caribbean Sea 3 m depth Phylogeny 9.431451° N; 82.356563° W Sporolithon molle C A92529 El Tor, Egypt, Gulf of Suez A. Kaiser, no date, no habitat data –––KY994121 – Lectotype of Sporolithon ptychoides f. mollis Sporolithon molle NCU 606657 El Tor, Egypt, Gulf of Suez T. Sauvage, W. E. Schmidt, D. KY980432 KY994116 – KY994120 KY980427 (LAF 5848) 28°14.0800 N; 33°36.1740 E Gabriel, 8.v.2012, 1–2 m depth Phylogeny, distribution

(continued) TABLE 1. (continued)

GenBank accession no. ID no./reason for inclu- GENERA CORALLINE THE Taxa sion Locality Collector/date habitat LSU COI psbA rbcL UPA Sporolithon ptychoides TRH C19-3412 El Tor, Egypt, Gulf of Suez A. Kaiser, no date, no habitat data –––KY994119 – (Heydrich #12) Lectotype of S. ptychoides f. dura Sporolithon ptychoides NCU 606663 El Tor, Egypt, Gulf of Suez T. Sauvage, W. E. Schmidt, D. KY980431 – MF034542 KY994118 KY980426 (LAF 5846) 28°14.0800 N; 33°36.1740 E Gabriel, 8.v.2012, 1–2 m depth Phylogeny, distribution Sporolithon ptychoides NCU 606660 El Tor, Egypt, Gulf of Suez T. Sauvage, W. E. Schmidt, D. KY980430 KY994109 MF034541 KY994117 – (LAF 5875) 28°14.0800 N; 33°36.1740 E Gabriel, 8.v.2012, 1–2 m depth Phylogeny, distribution SPOROLITHON Sporolithon sp. PHYKOS 4623 Gulf of Chiriquı, near Mono 10.x.2011 ––MF034548 –– Phylogeny, Feliz, Panama, Paciﬁc 6–7 m depth distribution Ocean 8.03042° N; 82.87574° W Sporolithon sp. LAF 6956A Sackett Bank, NWGMx J. Richards & S. Fredericq, KY980437 – MF034549 KY994126 KY980429 Phylogeny, 28°38.00 N; 89°33.0280 W 7.ix.2014, 65–68 m depth

distribution AND Sporolithon sp. LAF 6970B Dry Tortugas Vicinity, J. Richards & S. Fredericq, KY980438 – MF034550 KY994127 – Phylogeny, SEGMx 10.ix.2014 (collected from site), HEYDRICHIA distribution 24°31.4940 N; 83°19.7930 W 15.xii.2014 (collected from microcosm), 69 m depth Sporolithon RB 600359 Abrolhos Continental Shelf, R.G. Bahia 27.x.2013, 28 m depth KY980435 KY994111 MF034545 KY994122 – yoneshigueae Paratype, phylogeny, Bahia, Brazil distribution 17°48035″ S; 38°1103″ W Sporolithon RB 600360 Abrolhos Continental Shelf R.G. Bahia 27.x.2013, 28 m depth KY980436 KY994112 MF034546 KY994123 – yoneshigueae Paratype, phylogeny, Bahia, Brazil distribution 17°48035″ S; 38°1103″ W 1047 1048 JOSEPH L. RICHARDS ET AL. respectively. Dredged specimens from the NWGMx and the RAxML-HPC2 program as implemented on the online SEGMx were collected according to the protocol of Richards server “The CIPRES Science Gateway V. 3.3” (Miller et al. et al. (2014, 2016; Table 1) and were either preserved directly 2010) with a GTR+I+G model of evolution and the partition in the field or subsequently collected from microcosms estab- scheme found by PartitionFinder, 1,000 topological searches lished with field-collected material. Two specimens of from random restarts, and 1,000 bootstrap replicates to assess Sporolithon yoneshigueae originating from Bahia State, Brazil, branch support. were obtained from the Smithsonian Institution in Washing- Evaluation of pair wise distance distribution for five mark- ton, D.C. and a specimen of Heydrichia cerasina was collected ers. The distribution of raw pair wise distances (i.e., diver- from Western Cape Province, South Africa. gence) was computed for the five markers utilized in the DNA sequencing of newly collected specimens. Markers chosen present study in order to evaluate their phylogenetic informa- for PCR included the chloroplast-encoded genes rbcL (en- tiveness within the Sporolithales ingroup. Raw pair wise dis- codes the large subunit of the enzyme ribulose-1,5-bispho- tances (number of bp differences) were calculated in R with sphate carboxylase/oxygenase), psbA (encodes the package “Ape” (Paradis et al. 2004; R Core Team 2014) and photosystem II reaction center protein D1 gene), and divided by the alignment length. Prior to doing so, align- “Universal Plastid Amplicon” (UPA, partial 23S rDNA), as ments were cropped at their 50 and 30 ends when missing well as the mitochondrial-encoded gene “cytochrome oxidase data were present, and some short sequences were removed subunit I” (COI), and the nuclear-encoded LSU (partial 28S as to not overinflate computed pair wise distances because of rDNA). Total DNA was extracted from specimens at the missing data. Short (296 bp) and medium length (578 bp) University of North Carolina at Chapel Hill following the pro- alignments of rbcL were each analyzed separately. tocol established by Hughey et al. (2001) and modified by Scanning electron microscopy. Portions of the thallus from Gabrielson et al. (2011). Amplification of rbcL and psbA (in silica gel-dried specimens were removed using a single-edged part) markers was performed at the University of North Caro- razor blade and forceps. Vertical fractures were performed lina at Chapel Hill following, respectively, Gabrielson et al. on crustose portions, whereas protuberances were sectioned (2011) and Sissini et al. (2014). Resulting PCR products were longitudinally using a new razor blade for each fracture. Sec- sent to the DNA Analysis Core Facility at the University of tions were mounted using liquid graphite and coated with North Carolina at Wilmington for sequencing as in Adey 10–16.5 nm of gold. Specimens were viewed using a Hitachi et al. (2015). Aliquots of extracted DNA were sent to the S-3000N SEM at a voltage of 15 or 20 kV, housed in the University of Louisiana at Lafayette where PCR and sequenc- Microscopy Center at UL Lafayette. Cell dimensions were ing for COI, LSU, and UPA, and rbcL and psbA (in part), measured from SEM micrographs following the protocols of were performed as in Richards et al. (2014, 2016). Irvine and Chamberlain (1994) and Adey et al. (2005). Ten DNA sequencing of type specimens. Specimens were received cells of each vegetative cell type (hypothallial, perithallial, on loan from the following herbaria: C, NY, and TRH. Speci- epithallial, and meristematic cells) and 10 reproductive struc- mens that consisted of multiple fragments were compared to tures were measured for each specimen, except where other- the protologue to identify the most representative fragment wise noted. Terminology follows Woelkerling (1988) and to choose for DNA extraction. Specimens were examined Adey et al. (2015). with a dissecting microscope and fragments were selected from the thallus portions with no epiphyte overgrowth or if epiphytes were present, they were removed. Total DNA was RESULTS extracted from all type specimens in an independent labora- tory at Hartnell College following the precautionary guideli- Historical type specimens and topotypes. Unique rbcL nes for historical specimens proposed by Hughey and sequences were obtained from the type specimens Gabrielson (2012); extraction and primers for amplification of four Sporolithon species, S. ptychoides, S. molle, S. di- of a short portion of rbcL followed Hernandez-Kantun et al. motum, and S. episporum (Fig. 1). The rbcL sequences (2016). Alignment and phylogenetic analyses. COI (661 bp), LSU obtained from the lectotype specimens of S. pty- (550 bp), psbA (863 bp), UPA (369 bp), and rbcL alignments choides f. dura (263 bp) and S. ptychoides f. mollis were constructed as in Richards et al. (2014, 2016). Three (296 bp) were each an exact match to this rbcL seg- alignments of varying sequence length were generated for ment of longer sequences newly generated from our rbcL, including an alignment of short sequences (263– collection of these morphotypes from the type local- 296 bp), an alignment of medium and long sequences (644– ity at El Tor, Egypt. The rbcL sequence of the holo- 1,387 bp) and an alignment of mixed short, medium, and type specimen of Archaeolithothamnion dimotum long sequences (263–1,387 bp). Publically available sequences of Sporolithon spp. and Heydrichia spp. were downloaded from (263 bp) was within the Sporolithon clade, but did GenBank (last accessed 20 January 2017) and included in the not match any existing or newly generated Sporoli- alignments, as well as those for several members of the Hapa- thon sequences. The rbcL sequence (296 bp) lidiales, Corallinales, and Rhodogorgonales (Table S1 in the obtained from the holotype specimen of Archae- Supporting Information) for phylogenetic context. Members olithothamnium episporum was an exact match over its of the Ceramiales that were downloaded from GenBank or entire length to our recently collected specimen of were newly generated (Table 1, Table S1) served as the out- group in all analyses. A concatenated alignment of LSU, COI, S. episporum from Caribbean Panama. Unfortunately, psbA, and rbcL (3,454 bp) was constructed using Sequence PCR amplification of the type specimen of Archae- Matrix 1.7.8 (Vaidya et al. 2011). PartitionFinder (Lanfear olithothamnion erythraeum was unsuccessful. et al. 2012) was used to determine the best partition scheme Phylogenetics. The results of the ML analyses with and model(s) of evolution as implemented by RAxML. For concatenated (Fig. 2) or single genes (Fig. 1, the concatenated data set, 10 data blocks were used by Parti- Figs. S1–S6 in the Supporting Information) showed tionFinder to find the best partitioning scheme using ‘search=all’; nine blocks consisted of a data block for each some variation with regard to the monophyly of codon position of each protein coding gene and a single data Sporolithon caused by the variable positioning of sis- block for LSU, respectively. ML analyses were performed with ter clade species assigned to this genus, namely THE CORALLINE GENERA SPOROLITHON AND HEYDRICHIA 1049

FIG. 1. Phylogeny based on ML analyses of short (264–296 bp), and medium and long (644–1,387 bp) rbcL sequences. Branch numbers indicate bootstrap values out of 1,000 replicates. Newly generated sequences shown in boldface. Stars indicate type specimens, diamonds indicate topotype specimens.

S. yoneshigueae and newly sequenced specimens from Sporolithon ptychoides formed a monophyletic clade the Gulf of Mexico. The clade formed by these two with sequences of the authentic S. ptychoides speci- taxa branches at the base of other Sporolithon spp. in mens (NCU 606660, NCU 606663), which were the concatenated analyses had good support (87 exact matches with the lectotype specimen. BS). However, in single gene analyses, except with Sequences of specimens identified as S. ptychoides LSU (Fig. S4), the positioning of this clade received from Hawaii (represented by COI, LSU and UPA) no support (psbA, COI, UPA; Figs. S3, S5, S6) and and Brazil (represented by rbcL and psbA) could not was paraphyletic with respect to the other Sporolithon be compared to each other due to a lack of corre- species in rbcL phylogenies (Fig. 1, Figs. S1 and S2). sponding markers. The rbcL and psbA sequences of Depending on the markers and concatenation, the topotype material of H. woelkerlingii from South support within Heydrichia was generally low to mod- Africa were distinct from the rbcL or psbA sequences erate, and most of its taxa lie on long branches. of the New Zealand specimen identified as H. woelk- The order Sporolithales was monophyletic in con- erlingii in the single (Fig. 1, Figs. S1–S3) and concatenation and when inferred from single gene psbA catenated gene analyses (Fig. 2). Heydrichia and COI analyses, but paraphyletic with rbcL and homalopasta formed a strongly supported clade with LSU analyses. Divergence analyses show that LSU is the South African H. woelkerlingii (Fig. 2, Fig. S3), the most conserved marker, followed by UPA, psbA, whereas H. cerasina comprised a clade with lower COI, and rbcL (Fig. S7 in the Supporting Informa- support (Figs. 1 and 2, Figs. S1–S3; BS = 56–68). tion; Tables S2–S7 in the Supporting Information). Based on the rbcL, psbA, UPA, COI, and LSU As indicated from pair wise distances (see Tables sequence data, as well as the morpho-anatomical S2–S6), sister clade species of Sporolithon characters, we herein emend the descriptions of (S. yoneshigueae and S. sp. [LAF 6956A, LAF 6970B]) S. ptychoides and S. molle, resurrect the name S. dimo- were as divergent from the remaining Sporolithon tum, and formally describe S. eltorensis sp. nov. spp. as they were from Heydrichia spp. Sporolithon Heydrich 1897a: 66. Sporolithon and Heydrichia names. ML analyses To the generic description provided by Verheij of the single gene alignments (Fig. 1, Figs. S1–S6) (1993), we add the postfertilization developmental and concatenated gene alignment (rbcL, psbA, COI, features of Bahia et al. (2015b), including fertilized LSU; Fig. 2) show that none of the sequences carpogonia that each produce short, 1–2 celled deposited in GenBank of specimens identified as gonimoblasts (a central fusion cell is absent) each 1050 JOSEPH L. RICHARDS ET AL.

FIG. 2. Phylogeny based on ML analyses of concatenated COI, LSU, psbA, and rbcL (3,454 bp). Branch numbers indicate bootstrap values out of 1,000 replicates. Newly generated sequences shown in boldface. Stars indicate type specimens, diamonds indicate topotype specimens. bearing terminal, oblong carposporangia, which are wide (Fig. 3, B and C). Perithallium with multiple found along the floor and walls of carposporangial fusions, secondary pit connections rarely observed conceptacles. (Fig. 3, D and E), with cells 7–11.5 lm long 9 5– Sporolithon ptychoides Heydrich 1897a: 67–69, figs. 6.4 lm wide. Epithallium comprised of a single 2–3; pl. III. layer of armored epithallial cells 1–1.8 lm Basionym: S. ptychoides f. dura Heydrich 1897a: 67– long 9 3–6.7 lm wide (Fig. 3F). Meristematic cells 69, figs. 2–3, pl. III. subisodiametric, or slightly longer or shorter than Lectotype: TRH C19-3412, Heydrich #12, El Tor, wide, 5–10.5 lm long 9 4.8–10.2 lm wide Egypt, leg. A. Kaiser. Designated by Woelkerling and (Fig. 3F). Townsend (in Woelkerling, 1988: 204–210) (figs. Reproduction: Of the two specimens collected, one 239, 243, 245, 247). was tetrasporangial (NCU 606660), the other nonre- DNA sequences: A 263 bp sequence of rbcL was productive (NCU 606663). Tetrasporangial compart- obtained from the lectotype specimen of S. pty- ments borne among a basal layer of elongated cells, choides. This sequence was an exact match over its unshed and buried after spore release (Fig. 4, A–C). entire length to two 1,387 bp rbcL sequences, each Tetrasporangia 45–52 lm long 9 24–37 lm wide generated from specimens collected at the type with triangular stalk cells 10–17 lm long 9 10– locality in the shallow subtidal (1–1.5 m deep) of 20 lm wide (N = 8; Fig. 4D). Paraphyses 3–4 celled. the Gulf of Suez at El Tor, Egypt. Other diagnostic Rosette cells not observed (tetrasporangial compart- sequences from these topotype specimens included ments were not present at the surface in the frag- psbA, UPA, COI, LSU (Table 1), and tufA (GenBank ments examined). accession = KU362143). Distribution: Confirmed only from the Gulf of The following characterization of S. ptychoides is Suez, at El Tor, Egypt. based only on our observations of field-collected Sporolithon molle (Heydrich) Heydrich 1897b: specimens. 416–417. Morphology and Habitat: Thallus non-geniculate, Basionym: S. ptychoides f. mollis Heydrich 1897a: with rounded protuberances; overgrowing coral on 67–69, pl. III, figs. 16, 18, 19. shallow fringing reefs and reef flats, 1–1.5 m deep Lectotype: C A92529, El Tor, Egypt, leg. A. Kaiser. (Fig. 3A). Designated by Verheij (1993) (189–190, figs. 13, 14). Anatomy: Hypothallium monomerous with rectan- Isolectotype: TRH C19-3419, Heydrich #11, El Tor, gular-shaped cells 18.75–52 lm long 9 3.5–9 lm Egypt. THE CORALLINE GENERA SPOROLITHON AND HEYDRICHIA 1051

FIG. 3. Sporolithon ptychoides, NCU 606663 (A, left, B–F), NCU 606660 (A, right). (A) Thallus habit. Scale bars 0.6 cm (left images), 0.3 cm (right images). (B) Vertical fracture of crustose portion of thallus; arrow indicates basal region, scale bar 380 lm. (C) Detail of area indicated by arrow in B showing monomerous hypothallium (bracket) and upswept tiers of perithallial ﬁlaments, scale bar 85 lm. (D) Perithallium with multiple cell fusions (arrows), scale bar 19 lm. (E) Detail of cell fusion (arrow) neighboring a secondary pit connection (circle arrow), scale bar 20 lm. (F) Armored epithallial cells lacking epithallial cell roofs (arrows), meristematic cells (M), and recently divided meristematic cell (double arrowhead), scale bar 14 lm.

FIG. 4. Sporolithon ptychoides NCU 606660. (A) Longitudinal section of protuberance showing layers of unshed tetrasporangial compartments, scale bar 450 lm. (B) Layer of overgrown tetrasporangial compartments borne among layer of elongate cells, scale bar 80 lm. (C) Detail of paraphyses with elongate cells (arrows) growing among tetrasporangial instead of tetrasporangial compartments; stalk cell (circle arrow), scale bar 23 lm. (D) Detail of remnants of apical pore plug of tetrasporangial compartment, scale bar 25 lm. 1052 JOSEPH L. RICHARDS ET AL.

FIG. 5. Sporolithon molle, NCU 606657. (A) Thallus habit, scale bar 1 cm. (B) Vertical fracture of crustose portion showing hypothallium (bracket) and upswept tiers of perithallial ﬁlaments, scale bar 80 lm. (C) Longitudinal section of protuberance showing radial construction, scale bar 0.8 mm. (D) Perithallium with cell fusions and secondary pit connections (arrows indicates location shown in E), scale bar 50 lm. (E) Magniﬁed view of cell fusions in the x-axis (circle arrows) and z- axis (*) and secondary pit connections in the x-axis (arrow) and z-axis (arrowhead), scale bar 23 lm. (F) Armored epithallial cells (arrows), some with intact epithallial cell roofs (circle arrows), and meristematic cells (“M”), scale bar 17 lm.

DNA sequences: A 296 bp rbcL sequence was Reproduction: One tetrasporangial specimen was obtained from the lectotype specimen of S. molle. observed. Tetrasporangial pores in surface view 6– This sequence was an exact match over its entire 10.8 lm diameter surrounded by 9–12 rosette cells length to a 694 bp sequence generated from a spec- (Fig. 6, A and B). Tetrasporangial compartments imen (NCU 606657) collected at the type locality, (Fig. 6, C–F) borne among a layer of elongate cells El Tor, Egypt in the Gulf of Suez. Other diagnostic (Fig. 6D, Fig. S8 in the Supporting Information), sequences from this topotype specimen included unshed and buried after spore release (Fig. 6, D UPA, COI, LSU (Table 1), and tufA (GenBank and F). Tetrasporangia 51–58.5 lm long 9 21– accession = KU362144). 41 lm wide (Fig. 6, C–F, Fig. S9 in the Supporting The following characterization of S. molle is based Information) with triangular stalk cells 10–18.75 lm only on our observations of field-collected speci- long 9 14–19.5 lm wide (Fig. 6, D–F). Paraphyses mens. 3–4 celled. Morphology and Habitat: Thallus non-geniculate, Distribution: Confirmed only from the Gulf of Suez with warty to more slender protuberances; growing at El Tor, Egypt. attached to the benthic substratum on shallow fring- Sporolithon dimotum (Foslie & M.Howe) Yam- ing reefs and reef flats, 1–1.5 m (Fig. 5A). aguishi-Tomita ex Wynne 1986: 2258. Anatomy: Hypothallium monomerous, with rectan- Basionym: Archaeolithothamnion dimotum Foslie & gular-shaped cells 16–30 lm long 9 3–6.5 lm wide Howe 1906: 128–129, pl. 80 (1), 87. (Fig. 5, B and C). Perithallium with multiple fusions Holotype: NY 00900043 (M. A. Howe 2667), Lemon and secondary pit connections (Fig. 5, D and E). Bay, Near Guanica, Puerto Rico, 27.vi.1903, leg. M. Perithallial cells 6.8–19.5 lm long 9 5.5–10.2 lm A. Howe. wide. Epithallium comprised of a single layer of Isotype: TRH C19-3380. armored epithallial cells 1.5–2.5 lm long 9 4.2– DNA sequences: A 263 bp rbcL sequence (Table 1) 7.6 lm wide (Fig. 5F). Meristematic cells subisodia- was obtained from the holotype specimen of S. metric to rectangular, 6.4–12 lm long 9 5–6.8 lm dimotum. This sequence is diagnostic for this species wide (Fig. 5F). and was not identical to any sequences generated in THE CORALLINE GENERA SPOROLITHON AND HEYDRICHIA 1053

FIG. 6. Sporolithon molle, NCU 606657. (A) Surface view of tetrasporangial sorus, scale bar 80 lm. (B) Surface view of two tetrasporangia showing rosette cells (arrows) surrounding the pore (*), scale bar 23 lm. (C) Layers of unshed tetrasporangial compartments (arrows), scale bar 130 lm. (D) Tetrasporangial compartments showing neighboring elongate cells (arrows), scale bar 50 lm. (E) Tetrasporangial compartment showing triangular stalk cell (circle pointer) and neighboring elongate cells (arrows), scale bar 24 lm. (F) Oblique fracture through tetrasporangial compartments showing partial views of neighboring cells (image gives the false impression that neighboring cells are not elongate), scale bar 100 lm.

this study or to any sequences in GenBank (last Sporolithon eltorensis sp. nov. J.Richards & accessed 20 January 2017). P.W.Gabrielson Distribution: Known only from the type locality. Holotype: NCU 606659, Gulf of Suez, El Tor, Egypt Sporolithon episporum (M.Howe) E.Y.Dawson 1960: (28°14.0800 N; 33°36.1740 E), 8.v.2012, on sand and 40, figs. 11, 12. coral bottom, 1–2 m deep, leg. Will Schmidt, Basionym: Archaeolithothamnium episporum Howe Daniela Gabriel & Thomas Sauvage. 1918: 2, pl. 1–6. Isotype: Specimen LAF 5767, Gulf of Aqaba, Holotype: NY 00900041 (Howe 6832), Point Toro, Dahab, Egypt (28°28.6540 N; 34°30.6980 E), 1–2m near Colon, Panama Canal Zone (Caribbean coast), deep, leg. Will Schmidt, Daniela Gabriel & Thomas 7.i.1910, leg. M. A. Howe. Sauvage. Paratypes: In USNM, Cat. No. 35298. (Howe Etymology: The specific epithet refers to the holo- 6840). type locality, El Tor, Egypt. DNA sequences: A 296 bp rbcL sequence was Diagnosis: The diagnostic characters of this species obtained from the holotype specimen of S. epispo- are the following DNA sequences from the holo- rum, specifically the individual piece identified as type: psbA and LSU; and from the isotype: psbA, the “technical type” by Howe (1918) in the proto- UPA, COI, LSU (Table 1), and tufA (GenBank logue, and was an exact match over its entire length accession = KU362142). Thus far, no morpho-anato- to a 1,362 bp rbcL sequence of a specimen collected mical features distinguish this species from other from Caribbean Panama (Table 1). Other diagnos- Sporolithon species. tic sequences from this field-collected specimen Morphology and Habitat: Thallus encrusting, included COI and psbA (Table 1), the latter was an smooth to warty growing over and consolidating exact match to a published sequence (Table S1). underlying substratum of sand and coral (Fig. 7A); Distribution: Caribbean Panama; Costa Rica (Bahia shallow subtidal zone, 2–6 m deep, along fringing et al. 2014a,b). reefs and reef flats. 1054 JOSEPH L. RICHARDS ET AL.

FIG. 7. Sporolithon eltorensis NCU 606659. (A) Thallus habit, scale bar 6 mm. (B) Vertical fracture, scale bar 340 lm. (C) Hypothallium (lower bracket) and perithallium (upper bracket), scale bar 110 lm. (D) Perithallium showing cell fusions in the x-axis (circle pointer) and z-axis (*) and secondary pit connections in the x-axis (white arrow) and z-axis (black arrow), scale bar 23.4 lm. (E) Meristematic cells (“M”) and section and surface views of epithallial cells (arrows), some with intact roofs (circle arrows), scale bar 16.8 lm. (F) Surface view of epithallium, scale bar 11 lm.

Vegetative Anatomy: Thallus construction monomer- Heydrichia cerasina Maneveldt & van der Merwe ous, with a non-coaxial hypothallium (Fig. 7, B and 2012: 11–21, figs. 1–27. C). Hypothallial cells rectangular-shaped, 19–35 lm Holotype: L 0821464, Western Cape, South Africa, long 9 3–5.5 lm wide (Fig. 7C). Perithallium with 16.vi.2010, leg. G.W. Maneveldt, E. van der Merwe, secondary pit connections and cell fusions in a ratio C. van Gass & O. van Gass. of 1:1. Perithalial cells 5.7–10.2 lm long 9 4.2– Isotypes: L 0821455-63 and UWC 10/144. 6.2 lm wide. Epithallium a single layer of armored DNA sequences: Isotype material sequenced: psbA, cells that are 1.8–2.6 lm long 9 4.2–6 lm wide rbcL, COI, LSU (Table 1). (Fig. 7, E and F). Meristematic cells subisodiametric Distribution: Known only from a 10 km stretch of or rectangular that are 3.4–6 lm long 9 6.2–9 lm coast from Struisbaai to L’Agulhas, Western Cape wide (Fig. 7E). Province, South Africa. Reproduction: No reproductive structures were observed. Distribution: Gulf of Suez, near El, Tor Egypt and DISCUSSION Gulf of Aqaba, Dahab, Egypt. Sequencing type specimens of generitype species. Cur- Heydrichia woelkerlingii R.A.Townsend, Y.M.Cham- rently, the only method to unequivocally charac- berlain & Keats 1994: 177–186, figs. 1–28. terize the generitype species of coralline algae is Holotype: YMC 88/60, Oudekraal, Cape Peninsula, by DNA sequencing of the type specimen; this is Western Cape, South Africa, 6.vi.1986, leg. L. Ander- true for any coralline species, not just the generi- son. type species. Relying on morpho-anatomical char- Isotype: LTB. acterizations to identify many coralline taxa has DNA sequences: Topotype material sequenced: psbA resulted in frequent species misidentifications and, (GenBank accession = JQ917415) and rbcL (Gen- worse, polyphyletic genera (Gabrielson et al. 2011, Bank accession = KP142788; Table S1). Martone et al. 2012, Hind et al. 2014a,b, 2015, Distribution: Confirmed only from Western Cape 2016, Sissini et al. 2014, Adey et al. 2015, Hernan- Province, South Africa. dez-Kantun et al. 2015, 2016). Evidence for this THE CORALLINE GENERA SPOROLITHON AND HEYDRICHIA 1055 also abounds with species-distinct sequences pass- validated by comparison with type material. For this ing under single names in GenBank for numerous reason, we have provided a new name, S. eltorensis, geniculate and non-geniculate species (e.g., Amphi- for the third species of Sporolithon found at El Tor, roa beauvoisii, Spongites yendoi, Sporolithon durum), Egypt (Fig. 2, Figs. S3–S6). We believe it is preferable and as demonstrated herein for the generitype to name a new species based on this newly collected species of the only two genera in the order Sporo- material rather than designate that material as an epi- lithales, S. ptychoides and H. woelkerlingii. type for either S. erythraeum or S. crassum, as that We obtained a diagnostic 296 bp rbcL sequence would merely be guessing at their identities. from the lectotype specimen of Sporolithon ptychoides Sporolithon molle had been distinguished morpho- that enabled us to unequivocally apply this name to anatomically from Sporolithon ptychoides by the field-collected material from the type locality, El Tor, absence of a layer of elongate cells at the base of Egypt. DNA sequences from the field-collected mate- tetrasporangial compartments (Verheij 1993, Bahia rial allowed us to compare these with available Gen- et al. 2011, 2014a). However, we demonstrated that Bank sequences, including markers commonly used both S. molle and S. ptychoides exhibit this character. in phylogenetic analyses of corallines (rbcL, psbA, During sectioning, tetrasporangial compartments UPA, COI, and LSU). Specimens that were identified that have a three-dimensional tapered base (where as S. ptychoides based on morpho-anatomy and then the stalk cell is cut-off from the subtending cell lay- sequenced from Brazil (Bahia et al. 2014b), Hawaii ers), may partially overlap and obscure the sur- (Sherwood et al. 2010), and New Caledonia (Bittner rounding basal layer of cells. Furthermore, because et al. 2011) did not match any sequence from the the paraphyses may bend and curve as tetrasporan- authentic S. ptychoides from El Tor, Egypt (Fig. 1 and gial compartments develop and enlarge (Townsend 2, Figs. S3 and S4). Based on the several misapplica- et al. 1995), and/or due to oblique section angles tions of the name S. ptychoides to sequenced speci- (Kaewsuralikhit et al. 2012, herein), it is inherently mens from other ocean basins, we hypothesize that likely that incomplete views of basal cells will some- other specimens attributed to S. ptychoides based on times be observed in sections that may give the false morpho-anatomy from the Mediterranean Sea impression that these cells are not elongate. (Alongi et al. 1996), Indonesia (Verheij 1993), South Much of our current information about Africa (Keats and Chamberlain 1993), Gulf of Thai- Sporolithon ptychoides is based on specimens that we land and Andaman Sea (Kaewsuralikhit et al. 2012), now know are not S. ptychoides from DNA sequence and Brazil (Nunes et al. 2008, Bahia et al. 2011, Hen- comparison. Other reports of specimens believed to riques et al. 2014) are also likely different species. be that species based on their uninformative mor- Likewise, the taxon reported to be H. woelkerlingii pho-anatomical characters cannot be corroborated from New Zealand (Nelson et al. 2015), far from its currently. Notwithstanding our recent collections, type locality of the Western Cape Province, South we still are unable to adequately characterize S. pty- Africa, is not that species and likely represents an choides morpho-anatomically due to the absence of undescribed species. Material was identified as that any reproductive material, except buried tetraspo- species based on morpho-anatomy, but the DNA rangial compartments, and due to having only two sequences for both psbA and rbcL indicate a different field-collected specimens. The same is true for species (Figs. 1 and 2, Fig. S3). S. molle and S. eltorensis, for which we have only one Sporolithon species. There are four Sporolithon spe- and two field-collected specimens, respectively; how- cies whose type localities are El Tor, Egypt, S. pty- ever, having few specimens upon which to describe choides, S. molle, S. erythraeum, and S. crassum. Our a species is true for many non-geniculate coralline ability to obtain diagnostic rbcL sequences from the species as pointed out by Hernandez-Kantun et al. lectotype specimens of S. ptychoides and S. molle to (2016). This problem is exacerbated when multiple compare with field-collected topotype material at El species are passing under a single name, as is com- Tor, Egypt enabled us to unequivocally apply these mon with both geniculate (Hind and Saunders names. Unfortunately, we were unable to PCR 2013, Hind et al. 2015, 2016) and non-geniculate amplify DNA from the lectotype specimen (TRH (Sissini et al. 2014, Adey et al. 2015, Hernandez- C19-3438) of L. erythraeum (Rothpletz 1893), the old- Kantun et al. 2015) species. est epithet that could apply to Sporolithon species at Despite the absence of any distinguishing mor- this same type locality, El Tor, Egypt. Apparently, all pho-anatomical characters and sharing the same original material of S. crassum was destroyed (Woelk- type locality, Sporolithon molle, S. ptychoides, and S. el- erling 1988: 207), as also occurred for S. ptychoides torensis clearly are distinct species based on all mark- and S. molle, except for specimens sent to other her- ers that we sequenced (Figs. 1 and 2, Figs. S1–S5). baria. We have been unable to locate any original Our clarification of their molecular identity will material of S. crassum. In the future, original material facilitate further taxonomic reassessment and even- of S. crassum may be found and type material of S. ery- tual species description of specimens called “S. pty- thraeum may be sequenced with de novo technology, choides” as well as other Sporolithon species at other but until then, neither of these names should be locations. For example, reports of S. molle based applied anywhere in the world, as their use cannot be only on morpho-anatomy, from the Spermonde 1056 JOSEPH L. RICHARDS ET AL.

Archipelago, Indonesia (Verheij 1993), the Persian biogeographic provinces or ocean basins based on Gulf (De Clerck and Coppejans 1996, John and Al- morpho-anatomical characters (e.g., S. ptychoides, Thani 2014), and from Brazil (Bahia et al. 2014a) S. molle, and S. episporum) has been confirmed by all need to be reassessed. Reports of S. ptychoides DNA sequence data. Based on the current evidence, and other Sporolithon spp. fossils (Braga and Bassi there appears to have been much speciation in this 2007, Ghosh and Sarkar 2013), and the biogeo- genus with little concomitant morpho-anatomical graphical implications of such reports, need to be change, despite the antiquity of Sporolithales. re-evaluated considering the lack of diagnostic mor- Sporolithales generic relationships. Phylogenetic anal- pho-anatomical characters, especially considering yses of all of the data sets, particularly rbcL and psbA, the estimated age of the Sporolithales (136.4–130 for which we have more complete taxon sampling, million years B.P.; Aguirre et al. 2010, Yang et al. suggest that Sporolithon is paraphyletic and the rela- 2016). tionships among Heydrichia species are not well Sporolithon dimotum (type locality: Lemon Bay, Near resolved. In particular the relationship of S. yon- Guanica, Puerto Rico) was placed in synonymy with ishigue from Brazil and an undescribed species from Sporolithon ptychoides by Bahia et al. (2011) based on the Gulf of Mexico to other Sporolithon spp. needs sharing the following morpho-anatomical features of to be investigated in greater detail. It is likely that tetrasporangial specimens: (i) tetrasporangia DNA sequences from additional new world taxa will grouped in sori and raised above the thallus surface, help to resolve these relationships. (ii) presence of a basal layer of elongate cells in areas of developing tetrasporangia, (iii) tetrasporangial We are most grateful to the curators of the following herbaria paraphyses comprised of 3–5 cells, (iv) deeply buried for the loan of type material essential to this study: Nina tetrasporangial compartments, and (v) size of Lundholm (C), Barbara Thiers (NY), and Kristian Hassel (TRH). We thank the following collectors and providers of tetrasporangia. Despite sharing all of these features, a field-collected material: Daniela Gabriel, D. Wilson Freshwa- 263 bp rbcL sequence from a holotype specimen ter, Gavin W. Maneveldt, Kevin Miklasz, Ricardo G. Bahia, shows S. dimotum is a distinct species more closely Gabriel Johnson, and Juliet Brodie. Sincere thanks to Tom related to an undescribed species from Brazil than to Pesacreta and Mike Purpurea at the UL Lafayette Microscopy S. ptychoides (Fig. 1, Fig. S1). We await new field col- for help and advice while using the SEM. Wilson Freshwater, lections to obtain longer, more phylogenetically DNA Analysis Core Facility, University of North Carolina, Wilmington, and Talita Vieira-Pinto, University of S~ao Paulo, informative DNA sequences of different markers for provided sequencing support, and Todd Vision provided S. dimotum to confirm its evolutionary relationships. research space and equipment to PWG. SF acknowledges sup- Sporolithon episporum (type locality: Point Toro, near port from NSF grants DEB-0937978 (RedToL) and DEB- Colon, Panama Canal Zone, Caribbean coast) has 1455569 (ARTS); TS thanks the Graduate Student Organiza- been recognized as a distinct species since it was first tion and the Student Government Association at UL Lafayette described by Howe (1918). We confirm that the for assistance with travel expenses during the collection of sequenced specimen named S. episporum from Atlan- specimens in Egypt; PWG acknowledges a private family trust for research support. tic Costa Rica (Bahia et al. 2014b) is correctly identified. The application of this name from the Natal and Western Cape Provinces of South Africa needs to Adey, W. H., Chamberlain, Y. M. & Irvine, L. M. 2005. An SEM- be confirmed by DNA sequence data. Sporolithon sp. based analysis of the morphology, anatomy, and reproduction of Lithothamnion tophiforme (Esper) Unger (Corallinales, (PHYKOS 4623), collected from Gulf of Chiriquı, Rhodophyta), with a comparative study of associated North Pacific Panama, may correspond to Sporolithon howei Atlantic arctic/subarctic Melobesioideae. J. Phycol. 41:1010– (Me.Lemoine) N.Yamaguishi-Tomita ex M.J.Wynne 24. (type locality Coiba Island, Gulf of Chiriquı, Adey, W. H., Hernandez-Kantun, J. J., Johnson, G. & Gabrielson, Panama), or S. pacificum E.Y.Dawson (type locality P. W. 2015. DNA sequencing, anatomy, and calcification patterns support a monophyletic, subarctic, carbonate reef-form- Isla del Cano,~ Costa Rica). DNA sequencing needs to ing Clathromorphum (Hapalidiaceae, Corallinales, be performed on the holotype specimens of these Rhodophyta). J. Phycol. 51:189–203. species to determine the identity of PHYKOS 4623. Aguirre, J., Perfectti, F. & Braga, J. C. 2010. Integrating phy- There are 15 additional names of extant species of logeny, molecular clocks, and the fossil record in the evolution of coralline algae (Corallinales and Sporolithales, Sporolithon (Guiry and Guiry 2017) not treated Rhodophyta). Paleobiology 36:519–33. herein, and the type specimens of all of these need to Alongi, G., Cormaci, M. & Furnari, G. 1996. On the occurrence be sequenced. We already know that there are multi- of Sporolithon ptychoides Heydrich (Corallinales, Sporolitha- ple species passing under S. durum (Foslie) ceae, Rhodophyta) in the Mediterranean Sea. Cryptog. Algol. R.A.Townsend & Woelkerling based on DNA 17:131–7. Bahia, R. G., Amado-Filho, G. M. & Maneveldt, G. W. 2014a. Spor- sequence comparisons (Figs. S2 and S3). We do not olithon molle (Heydrich) Heydrich (Sporolithales, Corallino- know which morpho-anatomical characters, if any, phycidae, Rhodophyta): an addition to the Atlantic flora will be useful in segregating species of Sporolithon. To found on a remote oceanic island. Cryptog. Algol. 35:7–14. date, none of the characters previous considered to Bahia, R. G., Amado-Filho, G. M., Maneveldt, G. W., Adey, W. H., Johnson, G., Marins, B. V. & Longo, L. L. 2014b. Sporolithon be diagnostic for these species has been supported by tenue sp. nov. (Sporolithales, Corallinophycidae, Rhodo- DNA sequence data. Thus far, no species reported to phyta): a new rhodolith-forming species from the tropical have a widespread distribution across different southwestern Atlantic. Phycol. Res. 62:44–54. THE CORALLINE GENERA SPOROLITHON AND HEYDRICHIA 1057

Bahia, R. G., Amado-Filho, G. M., Maneveldt, G. W., Adey, W. H., Heydrich, F. 1897a. Corallinaceae, insbesondere Melobesieae. Ber. Johnson, G., Jesionek, M. B. & Longo, L. L. 2015a. Sporoli- Deutsch. Bot. Gesellschaft 15:34–71, pl. 3. thon yoneshigueae sp. nov. (Sporolithales, Corallinophycidae, Heydrich, F. 1897b. Melobesiae. Ber. Deutsch. Bot. Gesellschaft Rhodophyta), a new rhodolith-forming coralline alga from 15:403–20, pl. 18. the southwest Atlantic. Phytotaxa 224:140–58. Hind, K. R., Gabrielson, P. W., Jensen, C. P. & Martone, P. T. Bahia, R. G., Maneveldt, G. W., Amado-Filho, G. M. & Yoneshi- 2016. Crusticorallina gen. nov., a non-geniculate genus in the gue-Valentin, Y. 2015b. New diagnostic characters for the subfamily Corallinoideae (Corallinales, Rhodophyta). J. order Sporolithales (Corallinophycidae, Rhodophyta). J. Phy- Phycol. 52:929–41. col. 51:1137–46. Hind, K. R., Gabrielson, P. W., Lindstrom, S. C. & Martone, P. T. Bahia, R. G., Riosmena-Rodriguez, R., Maneveldt, G. W. & Amado 2014a. Misleading morphologies and the importance of Filho, G. M. 2011. Research note: first report of Sporolithon sequencing type specimens for resolving coralline taxonomy ptychoides (Sporolithales, Corallinophycidae, Rhodophyta) for (Corallinales, Rhodophyta): Pachyarthron cretaceum is Corallina the Atlantic Ocean. Phycological Res. 59:64–9. officinalis. J. Phycol. 50:760–4. Bailey, J. C. & Chapman, R. L. 1998. A phylogenetic study of the Hind, K. R., Gabrielson, P. W. & Saunders, G. W. 2014b. Molecu- Corallinales (Rhodophyta) based on nuclear small-subunit lar-assisted alpha taxonomy reveals pseudocryptic diversity rRNA gene sequences. J. Phycol. 34:692–705. among species of Bossiella (Corallinales, Rhodophyta) in the Bittner, L., Payri, C. E., Maneveldt, G. W., Couloux, A., Cruaud, eastern Pacific Ocean. Phycologia 53:443–56. C., De Reviers, B. & Le Gall, L. 2011. Evolutionary history of Hind, K. R., Miller, K. A., Young, M., Jensen, C., Gabrielson, P. the Corallinales (Corallinophycidae, Rhodophyta) inferred W. & Martone, P. T. 2015. Resolving cryptic species of Bos- from nuclear, plastidial and mitochondrial genomes. Mol. siella (Corallinales, Rhodophyta) using contemporary and Phylogenet. Evol. 61:697–713. historical DNA. Am. J. Bot. 102:1–19. Braga, J. C. & Bassi, D. 2007. Neogene history of Sporolithon Hey- Hind, K. R. & Saunders, G. W. 2013. A molecular phylogenetic drich (Corallinales, Rhodophyta) in the Mediterranean study of the tribe Corallineae (Corallinales, Rhodophyta) region. Palaeogeogr. Palaeoclim. Palaeoecol. 243:189–203. with an assessment of genus-level taxonomic features and Cabioch, J. & Boudouresque, C. F. 1992. Guide des algues des mers descriptions of novel genera. J. Phycol. 49:103–14. d’Europe. Delachaux & Niestle, Lausanne. Howe, M. A. 1918. On some fossil and recent Lithothamnieae of Chamberlain, Y. M. & Irvine, L. M. 1994. Lithophylloideae Setch- the Panama Canal Zone. US Natl. Mus. Bull. 103:1–3, pls. 1–11. ell. In Irvine, L. M. & Chamberlain, Y. M. [Eds.] Seaweeds of Hughey, J. R. & Gabrielson, P. W. 2012. Comment on “Acquiring the British Isles. Volume 1 Rhodophyta Part 2B Corallinales, DNA from dried archival red algae (Florideophyceae) for Hildenbrandiales. HMSO, London, pp. 58–112. the purpose of applying available names to contemporary De Clerck, O. & Coppejans, E. 1996. Marine algae of the Jubail genetic species: a critical assessment”. Botany 90:1191–4. Marine Wildlife Sanctuary, Saudi Arabia. In Krupp, F., Abuzi- Hughey, J. R., Silva, P. C. & Hommersand, M. H. 2001. Solving nada, I. A. & Nader, A. H. [Eds.] A Marine Wildlife Sanctuary for taxonomic and nomenclatural problems in Pacific Gigarti- the Arabian Gulf, Environmental Research and Conservation Follow- naceae (Rhodophyta) using DNA from type material. J. Phy- ing the 1991 Gulf War Oil Spill. NCWCD, Riyadh and Sencken- col. 37:1091–109. berg Research Institute, Frankfun a.M., Riyadh, pp. 199–289. Irvine, L. M. & Chamberlain, Y. M. 1994. Seaweeds of the British Farr, T., Broom, J., Hart, D., Neil, K. & Nelson, W. 2009. Common Isles. Vol. 1, Rhodophyta, Part 2B. Corallinales, Hildenbrandiales. coralline algae of northern New Zealand: an identification guide: The Natural History Museum, London, 276 pp. NIWA Information Series No. 70, NIWA Wellington, Ministry of Johansen, H. W. 1981. Coralline Algae, a First Synthesis. CRC Press, Fisheries project ZBD2004-007. Boca Raton, Florida, 239 pp. Foslie, M. 1900. Revised systematical survey of the Melobesieae. John, D. M. & Al-Thani, R. F. 2014. Benthic marine algae of the Det. Kong. Norske Vidensk. Selsk. Skr. 1900–5:1–22. Arabian Gulf: a critical review and analysis of distribution Foslie, M. & Howe, M. A. 1906. New American coralline algae. and diversity patterns. Nova Hedwigia 98:341–92. Bull. New York Bot. Garden 4:128–36, pls. 80–93. Kaewsuralikhit, C., Maneekat, S., Noiraksa, T., Patarajinda, S. & Gabrielson, P. W., Miller, K. A. & Martone, P. T. 2011. Morpho- Baba, M. 2012. First record of Sporolithon ptychoides Heydrich metric and molecular analyses confirm two distinct species of (Sporolithales, Corallinophycidae, Rhodophyta) from Thai- Calliarthron (Corallinales, Rhodophyta), a genus endemic to land. Cryptog. Algol. 33:265–76. the northeast Pacific. Phycologia 50:298–316. Keats, D. W. & Chamberlain, Y. M. 1993. Sporolithon ptychoides Hey- Ghosh, A. K. & Sarkar, S. 2013. Diversification of the family Spor- drich and S. episporum (Howe) Dawson: two crustose coral- olithaceae: a case of successful survival in the perspective of line red algae (Corallinales, Sporolithaceae) in South Africa. cretaceous–tertiary mass extinctions in India. Natl. Acad. Sci. S. Afr. J. Bot. 59:541–50. Lett. 36:215–24. Lanfear, R., Calcott, B., Ho, S. Y. W. & Guindon, S. 2012. Parti- Guiry, M. D. & Guiry, G. M. 2017. AlgaeBase. World-wide elec- tionFinder: combined selection of partitioning schemes and tronic publication, National University of Ireland, Galway. substitution models for phylogenetic Analyses. Mol. Biol. Evol. Available at: http://www.algaebase.org (last accessed 20 Jan- 29:1695–701. uary 2017). Maneveldt, G. W. & van der Merwe, E. 2012. Heydrichia cerasina Henriques, M. C., Coutinho, L. M., Riosmena-Rodrıguez, R., Bar- sp. nov. (Sporolithales, Corallinophycidae, Rhodophyta) ros-Barreto, M. B., Khader, S. & Figueiredo, M. A. O. 2014. from the southernmost tip of Africa. Phycologia 51:11–21. Three deep water species of Sporolithon (Sporolithales, Rhodo- Martone, P. T., Lindstrom, S. C., Miller, K. A. & Gabrielson, P. phyta) from the Brazilian continental shelf, with the descrip- W. 2012. Chiharaea and Yamadaia (Corallinales, Rhodophyta) tion of Sporolithon elevatum sp. nov. Phytotaxa 190:320–30. represent reduced and recently derived articulated coralline Hernandez-Kantun, J. J., Gabrielson, P. W., Hughey, J. R., Pez- morphologies. J. Phycol. 48:859–68. zolesi, L., Rindi, F., Robinson, N. M., Pena,~ V., Riosmena- Mateo-Cid, L. E., Gonzalez, A. C. M. & Gabrielson, P. W. 2014. Rodriguez, R. Le., Gall, L. & Adey, W. H. 2016. Reassessment Neogoniolithon (Corallinales, Rhodophyta) on the Atlantic of branched Lithophyllum spp. (Corallinales, Rhodophyta) in coast of Mexico, including N. siankanensis sp. nov. Phytotaxa the Caribbean Sea with global implications. Phycologia 190:64–93. 55:609–35. Miller, M. A., Pfeiffer, W. & Schwartz, T. 2010. Creating the CIPRES Hernandez-Kantun, J. J., Rindi, F., Adey, W. H., Heesch, S., Pena,~ Science Gateway for inference of large phylogenetic trees in Proceed- V., Le Gall, L. & Gabrielson, P. W. 2015. Sequencing type ings of the Gateway Computing Environments Workshop (GCE), 14 material resolves the identity and distribution of the generit- Nov. 2010. New Orleans, LA, pp. 1–8. rype Lithophyllum incrustans, and related European species Nelson, W. A., Sutherland, J. E., Farr, T. J., Hart, D. R., Neill, K. L. hibernicum and L. bathyporum (Corallinales, Rhodophyta). T., Kim, H. J. & Yoon, H. S. 2015. Multi-gene phylogenetic J. Phycol. 51:791–807. analyses of New Zealand coralline algae: Corallinapetra 1058 JOSEPH L. RICHARDS ET AL.

novaezelandiae gen. et sp. nov. and recognition of the Hapa- lidiales ord. nov. J. Phycol. 51:454–68. Supporting Information Nunes, J. M. C., Guimar~aes, S. M. P. B., Donnangelo, A., Farias, J. & Horta, P. A. 2008. Taxonomic aspects of three species of Additional Supporting Information may be non geniculate coralline algae from Bahia State, Brazil. Rodri- found in the online version of this article at the guesia 59:75–86. publisher’s web site: Paradis, E., Claude, J. & Strimmer, K. 2004. APE: analyses of phylogenetics and evolution in R language. Bioinformatics Figure S1. Phylogram based on ML analyses of 20:289–90. short rbcL sequences (263–296 bp). Branch num- R Core Team. 2014. R: A language and Environment for Statisti- cal Computing. R Foundation for Statistical Computing, bers indicate bootstrap values out of 1,000 repli- Vienna, Australia. Available at: www.R-project.org (last cates. Newly generated sequences shown in accessed 30 January 2016), boldface. Stars indicate type specimens, diamonds Richards, J. L., Gabrielson, P. W. & Fredericq, S. 2014. New indicate topotype specimens. insights into the genus Lithophyllum (Lithophylloideae; Coral- linaceae, Corallinales) from offshore the NW Gulf of Mexico. Figure S2. Phylogeny based on ML analysis of Phytotaxa 190:162–75. medium and long rbcL sequences (694–1,387 bp). Richards, J. L., Vieira-Pinto, T., Schmidt, W. E., Sauvage, T., Branch numbers indicate bootstrap values out of Gabrielson, P. W., Oliveira, M. C. & Fredericq, S. 2016. Molecular and morphological diversity of Lithothamnion spp. 1,000 replicates. Newly generated sequences (Hapalidiales, Rhodophyta) from deepwater rhodolith beds shown in boldface. Stars indicate type specimens, in the Northwestern Gulf of Mexico. Phytotaxa 278:81–114. diamonds indicate topotype specimens. Rösler, A., Perfectti, F., Peña, V. & Braga, J. C. 2016. Phylogenetic relationships of Corallinaceae (Corallinales, Rhodophyta): Figure S3. Phylogram based on ML analyses of taxonomic implications for reef-building corallines. Journal of psbA (863 bp). Branch numbers indicate bootstrap Phycology 52:412–431. values out of 1,000 replicates. Newly generated Rothpletz, A. 1893. Uber€ eine neue Pflanze (Lithothamnium erythraeum n.sp.) des Rothen Meeres. Bot. Zentralbl. 54:5–6. sequences shown in boldface. Stars indicate type Sherwood, A., Kurihara, A., Conklin, K., Sauvage, T. & Presting, specimens, diamonds indicate topotype specimens. G. 2010. The Hawaiian Rhodophyta Biodiversity Survey (2006-2010): a summary of principal findings. BMC Plant Figure S4. Phylogram based on ML analyses of Biol. 10:258. LSU (550 bp). Branch numbers indicate bootstrap Sissini, M. N., Oliveira, M. C., Gabrielson, P. W., Robinson, N. M., values out of 1,000 replicates. Newly generated Okolodkov, Y. B., Riosmena-Rodriguez, R. & Horta, P. A. sequences shown in boldface. Stars indicate type 2014. Mesophyllum erubescens (Corallinales, Rhodophyta)—so many species in one epithet. Phytotaxa 190:299–319. specimens, diamonds indicate topotype specimens. Thiers, B. 2017. (continuously updated) Index Herbariorum: A Figure S5. Phylogram based on ML analyses of global directory of public herbaria and associated staff. New York Botanical Garden’s Virtual Herbarium. Available COI (661 bp). Branch numbers indicate boot- at: http://sweetgum.nybg.org/ih/ (last accessed 08 May strap values out of 1,000 replicates. Newly gener- 2017). ated sequences shown in boldface. Stars indicate Townsend, R. A., Chamberlain, Y. M. & Keats, D. W. 1994. Hey- type specimens, diamonds indicate topotype spec- drichia woelkerlingii gen. et sp. nov., a newly discovered non- geniculate red alga (Corallinales, Rhodophyta) from Cape imens. Province, South Africa. Phycologia 33:177–86. Figure S6. Phylogram based on ML analyses of Townsend, R. A., Woelkerling, W. J., Harvey, A. S. & Borowitzka, M. 1995. An account of the red algal genus Sporolithon (Spor- UPA (369 bp). Branch numbers indicate boot- olithaceae, Corallinales) in southern Australia. Aust. Syst. Bot. strap values out of 1,000 replicates. Newly gener- 8:85–121. ated sequences shown in boldface. Stars indicate Vaidya, G., Lohman, D. J. & Meier, R. 2011. SequenceMatrix: con- type specimens, diamonds indicate topotype spec- catenation software for the fast assembly of multi-gene data- imens. sets with character set and codon information. Cladistics 27:171–80. Figure S7. Distribution of raw pair wise dis- Verheij, E. 1993. The genus Sporolithon (Sporolithaceae fam. nov., tances for each of the five markers analyzed in Corallinales, Rhodophyta) from the Spermonde Archipelago, Indonesia. Phycologia 32:184–96. this study. Divergence values for rbcL presented Woelkerling, W. J. 1988. The Coralline Red Algae: An Analysis of the for both short (296 bp) and medium length Genera and Subfamilies of Nongeniculate Corallinaceae. British (578 bp) alignments. Museum (Natural History) & Oxford University Press, Lon- don & Oxford, 268 pp. Figure S8. Sporolithon molle, NCU 606657. Woelkerling, W. J. & Townsend, R. A. 1988. Pgs. 203–210 in Unshed tetrasporangial compartments showing Woelkerling, W. J. 1988. The Coralline Red Algae: An Analysis layer of neighboring elongate basal cells (arrows). of the Genera and Subfamilies of Nongeniculate Corallinaceae. Bri- tish Museum (Natural History) & Oxford University Press, Image shows the same tetrasporangial compart- London & Oxford, pp. 268. ments as Figure 6D in the mirroring section of Wynne, M. J. 1986. A checklist of benthic marine algae of the the same fragment. tropical and subtropical western Atlantic. Can. J. Bot. 64:2239–81. Figure S9. Sporolithon molle, NCU 606657. Layer Yang, E. C., Boo, S. M., Bhattacharya, D., Saunders, G. W., Knoll, of unshed tetrasporangial compartments showing A. H., Fredericq, S., Graf, L. & Yoon, H. S. 2016. Divergence one with intact tetrasporangium and apical pore time estimates and the evolution of major lineages in the florideophyte red algae. Sci. Rep. 6:21361. plug (arrow). THE CORALLINE GENERA SPOROLITHON AND HEYDRICHIA 1059

Table S1. List of GenBank numbers for taxa Table S5. Divergence values (as %) for LSU included in analyses. Sequences included in the (446 bp). concatenated analyses shown in italics. N.A. = Table S6. Divergence values (as %) for COI data not available. *=sequence not analyzed in (661 bp). present study. Table S7. Divergence values (as %) for UPA Table S2. Divergence values (as %) for rbcL (368 bp). (296 bp). Table S3. Divergence values (as %) for rbcL (578 bp). Table S4. Divergence values (as %) for psbA (862 bp).