Ectopleura crocea ( L. Agassiz, 1862 )

Imazu, Maurício Antunes, Ale, Ezequiel, Genzano, Gabriel Nestor & Marques, Antonio Carlos, 2014, A comparative study of populations of Ectopleura crocea and Ectopleura ralphi (Hydrozoa, Tubulariidae) from the Southwestern Atlantic Ocean, Zootaxa 3753 (5), pp. 421-439 : 424-435

Ectopleura crocea ( L. Agassiz, 1862)

( Figures 3–5 ; Tables 1–8 )

Parypha crocea L. Agassiz, 1862: 249 , pls 23-2a.

Tubularia mesembryanthemum Allman 1871: 418 , figs. 83–84; Hargitt 1927: 494; Yamada 1959: 16; Schmidt 1971: 32, pl. 2B; Hirohito 1988: 18, fig. 4, pl.1, fig. B.

Tubularia crocea ; Torrey 1902: 42, pl. 3, figs. 22–23; Rees 1963: 1223; Brinckmann-Voss 1970: 28, text-fig. 30–34; Calder 1971: 24, pl. 1C; Genzano, Cuartas & Excoffon 1991: 69, pl. 5C; Blanco 1994: 182; Genzano & Zamponi 2003: 306, 307, 309, Tables 2-3 ; Demicheli & Scarabino 2006: 530.

Tubularia ralphi Bale 1884: 42 ; Watson 1980: 60, fig. 25–37; Watson 1982: 85, fig. 4.6.b, Plate 7.5.

Tubularia gracilis von Lendenfeld 1885: 597, fig. 51–52.

Tubularia sagamina Stechow 1907: 194 ; Yamada 1959: 16.

Tubularia australis Stechow 1925: 196 .

Tubularia warreni Ewer 1953: 351 , text-fig. 1–4; Millard 1959: 299; Millard 1966: 435; Millard 1975: 35, frontispiece, fig. 15A–G.

Ectopleura warreni ; Migotto & Silveira 1987: 101, fig. 3; Migotto 1996: 25; Grohmann et al. 1997: 230, Table 1 ; Rosso & Marques 1997: 417, 420, 421, Table 1 , Figure 4 .

Ectopleura crocea ; Petersen 1990: 174, fig. 27; Schuchert 1996: 107, figs. 64a–g; Schuchert 2001: 43; Bouillon et al. 2004: 104, fig. 55E–F; Genzano et al. 2009: 37, 40, Tables 2-3 , Fig. 3 ; Schuchert 2010: 357 -362, fig. 6.

Ectopleura ralphi ; Petersen 1990: 175; Schuchert 1996: 109; Migotto, Marques & Flynn 2001: 289, 290, 293-297, figs. 5-6, Table 1 .

Pinauay ralphi ; Marques & Migotto 2001: 475, 478, 480, figs. 2B, 3, Table 1 ; Migotto, Marques, Morandini & Silveira 2002: 10; Marques & Migotto 1994: 173, 174, Tables 15.1, 15.2; Grohmann 2006: 103, 104, Tables 1-2 ; Oliveira, Marques & Migotto, 2006: 4, 10, fig. 5, Table 1 ; Oliveira & Marques 2007: 31; Grohmann 2007: page not numbered; Silveira & Morandini 2011: 5.

Pinauay crocea ; Marques & Migotto 2001:480.

Examined material. Brazil: State of Rio de Janeiro, Macaé, Cavaleiros Beach, 22°24’S 41°47’W, 15.viii.2008, on rock, 95% ethanol, coll. A.C. Morandini ( MZUSP 1633); State of São Paulo: São Vicente, Vacas Beach, 23o58S 46o23W, 05.ix.1991, intertidal fringe, on rock, under Phragmatopoma stripe, at the same level of Eudendrium , 4% formalin, coll. A.C. Marques ( MZUSP 414); Itanhaém, Saudade Beach, 24o10'S 46o45'W, 26.viii.1991, intertidal fringe, on rock, under Phragmatopoma stripe, 4% formalin, coll. A.C. Marques ( MZUSP 406); Peruíbe, Centro Beach, 24o19'S 46o58'W, 12.viii.1992, intertidal fringe, on rock, sheltered place, forming an abundant stripe below Phragmatopoma , 18o C, 28‰, 4% formalin, coll. A.C. Marques ( MZUSP 433); Peruíbe, Jureia-Itatins Ecological Station, 24°34’S 47°14’W, 15.ix.2008, on rock, 95% ethanol, coll. J.M.M. Nogueira ( MZUSP 1636); Cananéia, Argolão Rocky Shore near São João Hill, 25o00'S 47o57'W, 25.viii.1992, intertidal fringe, on rock, 19o C, 28‰, 4% formalin, coll. A.C. Marques ( MZUSP 444); State of Paraná: Mel Island, Encantadas Rock, 25o34'S 48o18'W, 06.viii.1988, 4% formalin, coll. M.A. Haddad ( MZUSP 1750); Paranaguá, Yacht Club Paranaguá, 25°30’S 48°29’W, 10.x.2007, on artificial substrate, 95% ethanol, coll. M.A. Haddad, ( MZUSP 1637); Guaratuba, bottom trawling net, 4 km of shore, 25o52'S 48o33'W, 01.xii.2003, 4% formalin, coll. M.A. Haddad ( MZUSP 1751); State of Santa Catarina: Itapoá, Itapema beach, 26o05'S 48o36'W, 04.vi.2004, 4% formalin, coll. M.A. Haddad ( MZUSP 1752); Penha, on a culture of mussels, 26o45'S 48o38'W, 17.vi.2005, 4% formalin, coll. M.A. Haddad ( MZUSP 1753); Bombas, Bombas Beach, 27o07’S 48o30W, 03.xii.2006, 4% formalin, coll. M.A. Imazu ( MZUSP 1754) and 95% ethanol, coll. E. Ale, ( MZUSP 1638). Argentina: Mar del Plata, Punta Cantera, 38o04'S 57o32'W, 26.i.2002, intertidal fringe, 4% formalin, coll. G. Genzano ( MZUSP 1755), 95% ethanol, coll. G. Genzano ( MZUSP 1639).

Type specimens: Ectopleura ralphi , type specimen lost, a neotype was proposed based on specimens from Australia, Victoria, Port Phillip, Yarra River Entrance Beacon, 03.iv.1977, 1– 2m, on mussel and ascidia, formaldehyde ( NMV G3227) ( Watson 1980). Ectopleura crocea , we found no reference concerning the material described by L. Agassiz, from the port of Boston. It may be lost.

Description. Colonies dioecious, up to 55 mm high. Hydrorhiza and hydrocaulus with well-developed perisarc. Unbranched erect hydrocauli arising from stolonial hydrorhiza. Hydrocaulus’ coenosarc split into two longitudinal chambers with basal diameter 200–420 µm, apical 340–1000 µm; distal region of hydrocaulus with globular expansion supporting terminal hydranth. Hydranth with one whorl of aboral and one whorl of oral tentacles; oral tentacles adnate to hypostome up to the mouth region, circular in transversal section; aboral tentacles quadrangular in transversal section. Unbranched blastostyles of gonosomes arising immediately above aboral whorl of tentacles; main axis of each blastostyle supporting gonophores. Female gonophore cryptomedusoid, oval, with eight distal laterally compressed crests surrounding terminal aperture, terminal region of spadix projecting to outside. Male gonophore cryptomedusoid, spherical to oval, without distal crests. Early released actinulae with 8– 11 aboral capitate tentacles.

Aboral tentacles with four types of nematocysts: O-basitrichous isorhizas, rare and not measured ( Figure 3 A); basitrichous isorhizas, common, 6.45–12.19 X 2.84–6.04 µm ( Figure 3 B); desmonemes, abundant, spherical to oval, 3.50–6.81 X 2.41–5.13 µm ( Figure 3 C); small stenoteles, abundant, 5.02–7.82 X 3.66–8.09 µm ( Figure 3 E). Oral tentacles with three types of nematocysts: O-basitrichous isorhiza, rare and not measured ( Figure 3 A); basitrichous isorhizas, rare, 7.06–14.73 X 2.87–7.1 µm; large stenoteles, abundant, 7.36–12.5 X 6.07–11.38 µm and small stenoteles 4.93–7.93 X 3. 2–7.13 µm ( Figure 3 D).

Distribution in the South Western Atlantic Ocean. Brazil: States of Espírito Santo ( Grohmann et al. 1997, Grohmann 2006), São Paulo (Migotto & da Silveira 1987, Migotto 1996, Rosso & Marques 1997, Migotto et al. 2001, 2002, Marques & Migotto 2004, Oliveira et al. 2006, Oliveira & Marques 2007, Silveira & Morandini 2011), Paraná (Haddad, 1992), Santa Catarina ( Miranda et al. 2011) and Rio Grande do Sul ( Migotto & Silveira 1987) (see Migotto et al. 2002, Marques et al. 2003). Uruguay ( Demicheli & Scarabino 2006). Argentina: Provinces of Buenos Aires ( Blanco 1994, Genzano et al. 1991, 2009, Genzano 1994, 1998, Genzano & Rodriguez 1998), Río Negro and Chubut ( Blanco 1994, Genzano et al. 1991), Santa Cruz and Tierra del Fuego (Oliveira et al. submitted).

Remarks. General morphology. The species E. crocea and E. ralphi are morphologically similar and have been considered sister-taxa (Marques & Migotto, 2001) or synonyms ( Bouillon et al. 2006, Schuchert 2010). Historically, subtle differences have been cited to differentiate the species ( Table 1 ). For instance, Petersen (1990) differentiated the two species by describing E. crocea as having more aboral and oral tentacles than E. ralphi . However, this relationship appears to vary ( Table 1 ). Indeed, characters related to the tentacles are generally variable ( Tables 1–2 ; see also Agassiz 1862, Hargitt 1927, Ewer 1953, Calder 1971, Schmidt 1971, Migotto & Silveira 1987, Hirohito 1988).

The SWAO specimens all have hydroucauli that broaden distally ( Table 3), as described for E. crocea ( Hirohito 1988, Schuchert 1996), although specimens with the same diameter throughout the hydrocaulus were reported by Petersen (1990). Each polyp is gonochoristic, although settlement of actinulae on already developed hydrocauli ( Rungger 1969) may promote pseudo-hermaphroditism, a strategy also reported for other hydroids ( Brinkmann-Voss 1970, Sommer 1990, Marques 2001, Schuchert 2010, Nawrocki & Cartwright 2012).

There are inconsistencies in the description of the blastostyles, either characterized as unbranched ( Ewer 1953 and Millard 1966 for E. ralphi ; Brinckmann-Voss 1970, Hirohito 1988, and Petersen 1990 for E. crocea ) or branched ( Bale 1884, Millard 1975, Watson 1980, Migotto & Silveira 1987, and Petersen 1990 for E. ralphi ; Schuchert 1996 for E. crocea ) ( Table 4). Descriptions of the neotype of E. ralphi from Melbourne ( Australia) have different ways to describe the blastostyles, characterized either as “[...] only occasionally branched” ( Schuchert 1996: 109; our underline) or as “mature blastostyles branched [...]” ( Watson 1980: 61), or (“usually unbranched, but some branching can occur”, Schuchert 2010, p. 359). Specimens from SWAO have unbranched blastostyles ( Figure 4 A–B), similar to those described by Allman (1871) and contrasting with the long bunches of gonophores described for other localities ( Calder 1971, Millard 1975, Watson 1982, Petersen 1990). Millard (1959) commented on the difficulty of differentiating primary and secondary pedicels, which may explain the challenges of characterizing the ramification of blastostyles.

(shaded rows), expressed as minimum–maximum values. Length and diameter are given in millimeters. T = Tubularia ; E

= Ectopleura .

Hydranth Tentacles

Locality length diameter aboral oral

(n) number length number length Hydrocaulus Gonophores

Locality length diameter length diameter (n) proximal medial distal

São Vicente 14–35 240–300 300–400 340–540 260–500 180–300 (10) (20.7±6.17) (280±23.09) (350±28.67) (436±65.18) (368±68.77) (228±39.1) Author Female Male Blastostyles Actinulae (Original reference to species) gonophore gonophore

Agassiz (1862) (Paripha 6–10 crests no crests simples or branched

crocea )

Allman (1871) 8 apical processes 4 small tubercles no oral tentacles

( T. mesembryanthemum )

Bale (1884) ( T. ralphi ) 4tubercles branched

Allen (1900) ( Paripha crocea ) 6–8 crests no crests branched

Torrey (1902) ( T. crocea ) 8 crests top spherical, smooth or

with processes

Stechow (1907) ( T. sagamina ) 8 tubercles 6 tentacless

Stechow (1925) ( T. australis ) 4 small tubercles 8 aboral, 5 oral tentacles

Hargitt (1927)

( T. mesembryanthemum )

The morphology of the female gonophores ( Figure 4 C) is a diagnostic feature for E. crocea ( Rees & Thursfield 1965, Schmidt 1971, Petersen 1990, Schuchert 2001, 2010). Nevertheless, some variation may be observed, mainly because of the development of the gonophores, or due to their contraction ( Torrey 1902, Schuchert 1996). It is hard to identify tubulariids without gonophores ( Watson 1982), and this may be the cause of misidentifications between E. crocea and Ectopleura larynx (Ellis & Solander, 1786) , especially for the Northwestern Atlantic (e.g. Fraser 1944).

Male gonophores do not present apical crests ( Figure 4 B; see also Agassiz 1862, Brinckmann-Voss 1970, Calder 1971, and Schuchert 1996, 2010 for E. crocea ; Ewer 1953, Watson 1980, 1982, Migotto & Silveira 1987 for E. ralphi ), although in some cases they are described with small apical processes, varying in size and development between colonies, and even in the same colony ( Table 4) ( Allman 1871, Hirohito 1988, and Petersen 1990 for E. crocea ; Stechow 1925 and Millard 1975 for E. ralphi ).

Larval characters, such as tentacles, were already used to separate E. ralphi and E. crocea . A vague note in Schuchert (1996: 109, appears to refer to the observations of someone else) states that “the actinulae of E. ralphi , however, are reported to have rudiments of oral tentacles which are absent from E. crocea ”. Other data refer to variations in morphology and number of tentacles of the actinulae ( Ewer 1953, Brinckmann-Voss 1970, Millard 1975, Watson 1980, 1982, Migotto & Silveira 1987, Petersen 1990, Schuchert 1996, 2010; see Table 4). Presumably “the oral tentacles will develop anyway immediately after the release of the actinula and the presence or absence in liberated ones is thus only a matter of timing” ( Shuchert 2010, p. 361), an interpretation that attenuates the importance of the variation. The actinulae of the SWAO present 8–10 ( Brazil) or 8–11 ( Argentina) capitate aboral tentacles, depending on their development, and do not have rudiments of oral tentacles ( Figure 4 D– E), but the variability of this character was never strictly assessed.

Detailed studies on anatomical and histological characters have corroborated previous observations. Among these, the histological preparations confirmed that the oral tentacles are circular in a transversal section ( Figure 5 A–B), while the aboral tentacles are squared ( Figure 5 C), as described by Petersen (1990) for E. crocea . Also, transversal sections of the hydrocaulus have shown that its coenosarc is split into two longitudinal chambers ( Figure 5 D), as already noted before ( Ewer 1953 for E. ralphi ; Allman 1871 and Schuchert 1996 for E. crocea ), although this feature was considered to be inconstant in number and size ( Millard 1959, 1975 for E. ralphi ; Campbell & Campbell 1968, Hirohito 1988, and Petersen 1990 for E. crocea ).

Cnidome. The cnidome was uniform throughout all studied populations from SWAO ( Figures 3 A, B, D, E). Literature data for E. crocea present few discrepancies ( Table 5 , contrasting with Table 6 ), for instance a cnidome restricted to stenoteles and desmonemes ( Brinckmann-Voss 1970) and a potential contamination by microbasic euryteles ( Schuchert 1996; in Schuchert 2010, p. 360, they are referred to “rare euryteles”).

Schuchert (1996) length 8.0–9.5 9.0–9.5 6.5–7.0 5.0–6.0 5.0–5.5 9.0 ( Ectopleura crocea ) width 7.0–9.5 3.0–4.0 5.0–5.5 3.5–4.5 3.0–3.5 4.0

São Vicente Itanhaém Peruíbe Cananéia Ilha do Mel Basitrichous length 8.58–10.74 6.45–10.41 7.77–12.12 8.45–10.55 *1 8.14–11.86 (9.43±0.48) (8.39±0.67) (9.39±0.67) (9.41±0.46) (9.76±0.8) Stenoteles length 5.57–7.34 5.59–7.22 5.31–7.43 5.45–7.82 5.48–7.64 (6.29±0.32) (6.29±0.31) (6.36±0.37) (6.52±0.44) (6.45±0.41) (small) width 4.2–5.96 4.27–5.5 4.14–6.67 4.33–6.05 4.49–6.04 (4.93±0.31) (4.85±0.27) (5.07±0.4) (5.06±0.36) (5.27±0.32) Basitrichous length 7.06–11.6 7.67–10.21 7.22–10.78 8.2–9.03 *2 7.68–11.97 (9.32±0.74) (8.67±0.4) (9.46±0.63) (8.62±0.59) (9.79±0.79) width 2.87–5.24 3.19–4.81 3.04–4.94 3.75–4.09 *2 2.94–5.67 (3.95±0.46) (3.91±0.31) (3.91±0.37) (3.92±0.24) (4.46±0.49) Oral Stenoteles length 7.37–11.88 7.43–11 7.36–11.37 8.27–11.47 *3 7.99–11.72 (9.72±0.89) (9.53±0.58) (9.68±0.68) (10.2±0.77) (10.61±0.59) tentacles (large) width 6.07–10.29 6.45–9.86 6.53–10.4 6.21–10.31 *3 7.07–10.42 (8.36±0.86) (8.23±0.66) (8.81±0.71) (8.79±0.86) (9.37±0.6) Stenoteles length 5.22–7.31 5.34–6.96 5.52–7.93 5.66–7.2 *4 5.64–7.47 (6.19±0.4) (6.06±0.3) (6.29±0.49) (6.4±0.37) (6.42±0.38) (small) width 4.24–6.5 4.1–6.45 4.33–7.02 4.34–5.82 *4 4.41–6.46 (5.03±0.37) (4.84±0.32) (5.14±0.53) (5.06±0.34) (5.3±0.45)

Guaratuba Itapoá Penha Bombas Mar del Plata Basitrichous length 8.04–12.19 8.91–11.91 8.65–11.56 7.37–10.61 7.81–10.43 (9.94±0.75) (10.6±0.6) (10.12±0.58) (9.24±0.56) (9.37±0.52) width 3.06–5.87 3.73–5.82 3.37–5.29 3.54–5.33 3.17–5.26

(4.41±0.5) (4.69±0.4) (4.4±0.4) (4.28±0.37) (4.35±0.43)

Aboral Desmoneme length 3.94–6.81 4.57–6.66 4.51–6.6 4.67–6.54 4.02–6.17

(5.14±0.53) (5.46±0.39) (5.67±0.41) (5.68±0.44) (5.2±0.37)

tentacles width 2.53–4.49 3.2–5.13 3.2–4.9 3.2–4.85 3.08–4.36

(3.67±0.46) (4.14±0.33) (4.12±0.38) (4.08±0.31) (3.67±0.28) Stenotele length 5.02–7.3 5.77–7.31 5.37–7.66 5.13–7.69 5.26–6.96

(6.19±0.47) (6.62±0.3) (6.47±0.42) (6.4±0.42) (6.17±0.32) (small) width 3.66–5.69 4.83–8.09 4.3–6.06 4.27–6.04 4.24–5.78

(4.72±0.42) (5.49±0.47) (5.01±0.35) (5.0±0.37) (4.88±0.3) Basitrichous length 7.9–11.16 7.72–13.05 7.94–11.38 7.8–11.34 7.65–14.73 (9.38±0.7) (10.38±0.7) (9.7±0.77) (9.39±0.65) (9.32±0.94) width 2.87–5.18 3.51–5.56 3.42–5.52 3.41–5.73 3.28–7.1

(4.21±0.45) (4.55±0.47) (4.28±0.41) (4.23±0.4) (4.24±0.55)

Oral Stenotele length 8.64–11.87 8.19–12.15 9.25–12.5 8.98–12.22 8.68–12.09 (10.31±0.58) (10.29±0.84) (10.8±0.56) (10.3±0.63) (10.42±0.71)

tentacles (large) width 7.64–9.87 7.05–11.16 8.19–11.38 7.64–10.38 7.69–10.49 (8.67±0.56) (9.19±0.8) (9.76±0.61) (9.09±0.52) (8.98±0.59) Stenotele length 5.03–7.08 5.09–7.37 5.28–7.22 4.93–7.48 5.23–7.26

(6.12±0.41) (6.4±0.46) (6.32±0.38) (6.18±0.47) (6.19±0.37) (small) width 3.2–3.2–6.00 4.21–7.13 4.43–5.78 3.9–6.02 4.05–5.84

(4.6±0.43) (5.34±0.4) (5.07±0.32) (4.95±0.43) (4.96±0.38) The basitrichous isorhizas found in SWAO specimens were characterized for “ Tubularia larynx ” Ellis & Solander, 1786 as “pseudo-microbasic b-mastigophore” ( Östman et al. 1995: 166), possibly because of the basal spines of the tubule, giving the false appearance of a shaft under light microscopy (Östman 1987). The types isorhiza/anisorhiza and basitrichous/b-mastigophore are sometimes suggested as overlapped categories ( Cutress 1955, England 1991).

The cnidome has been suggested to be a valuable taxonomic tool, even in more restricted geographic scales (Östman et al. 1987), but for the SWAO region, neither cnidome, nor dimensions of the hydroid, present any kind of geographic structure. Specimens from the states of São Paulo ( São Vicente, Itanhaém, Peruíbe e Cananéia) and Santa Catarina (Itapoá and Penha) have the smallest and largest nematocysts, respectively ( Figure 6 A; Table 6 ). Comparing general dimensions of the hydroid ( Figure 6 B; Tables 2-3 ), specimens from Guaratuba and Mar del Plata have the lowest and highest values, respectively.

DNA. Twenty-two specimens were analyzed for each marker. ITS1+ 5.8S (332bp) resulted in a total of six haplotypes, with the Argentine haplotype identical to that of Santa Catarina, both basically with a distance of three base pair indel from all others ( Figure 7 A). The genetic diversity found among the six haplotypes was only 0.1% ( Table 7 ).

COI (489bp) resulted in a total of 11 haplotypes. Polymorphic sites revealed no saturation ( Table 7 ), and 20 substitutions, with one synonymous transversion and four non-synonymous transitions (two replacements of Valine per Isoleucine, both aliphatic/hydrophobic; two replacements of aromatic Phenilalanine per aliphatic Leucine, both hydrophobic). These five changes were detected in only five Brazilian haplotypes. The COI haplotype network also shows low genetic distances between haplotypes ( Figure 7 B). It is interesting to notice that the genetic pairwise difference between the most distant sites (northernmost Macaé and southernmost Mar del Plata) is lower than those among closer intermediate localities ( Table 8 ). Corrected average of pairwise differences and pairwise fixation indexes between localities showed no significant values ( Table 8 ). A recent publication included samples of Ectopleura from both South Africa (southern hemisphere, expected to be geographically more related to E. ralphi ) and California (northern hemisphere, expected to be geographically more related to E. crocea ), and found no genetic difference between these two populations (Nawrocki & Cartwright 2012).

1Total number of mutations detected

2Singleton variable sites

3Parsimony informative sites

4Number of sites with insertion/deletion of bases 5Relation between synonymous/non-synonymous amino acid changes 6Nucleotide diversity/standard deviation

General discussion and concluding remarks. We have observed no significant difference or geographical patterns between Brazilian E. ralphi and Argentine E. crocea for both morphological and molecular data. Therefore, SWAO populations of Ectopleura likely belong to the same species.

Another important question is whether both species are valid. Schuchert (1996: 109) remarked that “until further, preferably also molecular, analysis has been made it seems advisable to keep both species separate”. For Petersen (1990:176), “despite the obvious similarities between E. ralphi and E. crocea it seems best at present to keep the two species separate since the differences listed above appear to be constant and are shared by population of E. ralphi in both Australia and South Africa ”. Traditionally, the majority of the records of E. crocea and E. ralphi are from the Northern and Southern hemisphere, respectively ( Figure 1 ). The fragmented and disjunct distributions of the species could also be explained by bioinvasion processes, either human or naturally mediated ( Hewitt 2002, Ruiz et al. 2006, Marques, 2011, Mead et al. 2011, Rocha et al. 2013).

Considering our current knowledge of E. crocea and E. ralphi , we see no evidence to keep them apart. Considering them as synonymous, the binomen E. crocea would have nomenclatural priority. This was first proposed by Bouillon et al. (2006: 252) and reiterated by Schuchert (2010, that considered as a “new syn.”, p. 357). The proposal by Bouillon et al. (2006) offered no concrete evidence or arguments supporting the synonymy. In contrast, Schuchert (2010) proposed a long synonymic list for E. crocea , but his analysis was actually based on very few specimens (n=12) from four restricted regions ( Australia, South Africa, Mediterranean and Atlantic USA) – even in this restricted sample the phenotypic variability reported is impressive, which is also consistent with a composite of species (see Tables 1 and 4 for a summary). Schuchert (2010, p. 359) listed as diagnostic characters of E. crocea the “female sporosacs usually bearing six to eight crest-like processes around distal opening, several eggs or embryos per sporosac”.

In this study we provide substantial evidence that two different SWAO “populations”, previously assigned to E. crocea and E. ralphi , are the same. Nevertheless, whether these two different “populations” correctly represent the world nominal species E. crocea and E. ralphi is difficult to assess, and has to be considered conjectural. The proposed synonymy between E. crocea and E. ralphi ( Bouillon et al. 2006, Schuchert 2010) have not been based on strict taxonomical procedures, i.e., neither were based on the study of the type specimen of E. ralphi , nor on broad geographic analysis considering extensive phenotypic variation of abundant material of the species, nor on broad molecular analysis from a wide geographical range. Therefore, the ad hoc proposal of the synonymy, even though presently accepted by us, has to be considered tentative and subject to further assessment.

Figures

FIGURE 1. Worldwide distribution of Ectopleura crocea and Ectopleura ralphi (data from Agassiz 1862; Allman 1871; Allen 1900; Torrey 1902; Stechow 1907, 1925; Hargitt 1927; Ewer 1953; Millard 1959, 1966, 1975; Yamada 1959; Rees 1963; Brinckmann-Voss 1970; Calder 1971; Schmidt 1971; Watson 1980, 1982; Migotto & da Silveira 1987; Hirohito 1988; Petersen 1990; Schuchert 1996; Bouillon et al. 2004; Genzano 2005).

FIGURE 3. Different types of nematocysts documented under light microscopy. A: O-basitrichous isorhizas, B: basitrichous isorhizas, C: desmonemes, D: large stenoteles, E: small stenoteles.

FIGURE 4. Morphological traits of gonophores (A – D, scanning electronic microscopy) and actinulae (E, light microscopy). A: Oral-lateral view of the hydranth with its blastostyles of male gonophores. B: Detail of the blastostyles with male gonophores. C: Female gonophore with the apical crests on top. D: Female gonophore liberating an actinula. E: Liberated actinula with capitate aboral tentacles. Legends on figures: Gn, gonophore; Ac, actinulae; S, spadix.

FIGURE 5. Histological sections documented under light microscopy. A: transversal section at the base of oral tentacles, where they are still adnate to hypostome. B: transversal section of the oral tentacles showing their circular shape. C: oblique section of the aboral tentacles showing the squared section of the cells constituting that. D: transversal section at distal region of hydrocaulus showing its division into two longitudinal chambers. Legends on figures: HO, hypostome; OT, oral tentacle; AT, aboral tentacle; S, septum; C, chamber; P, perisarc.

FIGURE 6. Principal component analyses (PCA) for cnidome (A) and morphology (B). Note that the minimum and maximum cases in A (arrows) do not match with minimum and maximum cases in B. Most similar cases marked by dashed circles.

FIGURE 7. Haplotype parsimony networks for ITS 1 + 5.8 S (A) and COI (B) for the different populations of Ectopleura crocea and Ectopleura ralphi of the Southwestern Atlantic Ocean. Black circles for the Argentine haplotypes, white ones for Brazilian localities; the sizes of the circles are proportional to the number of localities sharing the haplotype; small black circles represent hypothetical or unsampled haplotypes. Lines connecting circles represent one substitution step. Note that A BO / MP samples differ from JU / PA / BO only by an indel of three base pairs (AAT / – – –). Locality codes are represented inside the circles as MA: Macaé, JU: Juréia, PA: Paraná, BO: Bombas, MP: Mar del Plata.

Tables

TABLE 1. Literature data of morphometric and meristic information for “ E. crocea ” (white rows) and “ E. ralphi ”

Author (Original reference to species)	Hydrocaulus length diameter	Aboral tentacles number length	Oral tentacles number length	Locality
Agassiz (1862) ( Paripha crocea )	63.5–88.9	24	24	Boston harbor, USA
Allman (1871) ( T. mesembryanthemum )	101	20–24	24	Gulf of Spezia, Italy
Bale (1884) ( T. ralphi )	76–101			Hobsons Bay, Australia
Allen (1900) ( Paripha crocea )	63.5–76.2	16–24		Woods Holl, USA
Torrey (1902) ( T. crocea )	inconstant	24 top	15–18	San Francisco and San Diego Bays, USA
Stechow (1907) ( T. sagamina )	150	50 9.0	20–25	Misaki, Japan
Stechow (1925) ( T. australis )	50–80	20	16–18	Fremantle, Australia
Hargitt (1927) ( T. mesembryanthemum )	35–50	20–25	20–25	South China
Ewer (1953) ( T. warreni )	100 top 0.22–0.8	24–30 4 top	22–27 0.8 top	Natal, South Africa
Millard (1959) ( T. warreni )	50	18–29	15–24	Natal and East coast, Africa
Yamada (1959) ( T. mesembryanthemum )	30	20	10	Sagami Bay and Seto, Japan; Amoy, South China
Yamada (1959) ( T. sagamina )	150	50	20–25	Sagami Bay and Seto, Japan; Amoy, South China
Rees (1963) ( T. crocea )	50			England
Millard (1966) ( T. warreni )	17.5 1.0			East cost of South Africa
Brinckmann-Voss (1970) ( T. crocea )		28 top	17–20	Italy, England and France
Calder (1971) ( T. crocea )	100	20–24	20–24	Florida and north of Gulf of Mexico
Schmidt (1971) ( T. mesembryanthemum )	50	20–25	20–25	Gulf Aqaba, Red Sea
Millard (1975) ( T. warreni )	50–100	31 top 5 or more	27 top 1 or more	Durban Harbor, South Africa
Watson (1980) ( T. ralphi )	120 top 0.3–0.5	16–27 4–5	15–25 2–3	Victoria, Australia
Watson (1982) ( T. ralphi )	120	16–27	15–25	Fremantle, Australia
Migotto & da Silveira (1987) ( E. warreni )	70 top 0.18–0.6	12–27 0.4–6.0	11–30 0.4–1.9	Southeastern and Southern coast, Brazil
Hirohito (1988) ( T. mesembryanthemum )	40–50	16–31	14–43	Japan
Petersen (1990) ( E. crocea )	70	22–30	18–24	East and West coast, USA
Petersen (1990) ( E. ralphi )		16–27	15–25	InkermanCreek,Australia
Schuchert (1996) ( E. crocea )	50 0.2–0.6	22 top	20	New Zealand
Bouillon et al. (2004) ( E. crocea )	70	22–30	18–24	Mediterranean
Schuchert (1996) ( E. crocea )	30–80, 0.6–0.8, maxi 120 but up 2	22–28 (max. 38)	ca. 18 (max. 26)	Australia, South Africa, USA (Atlantic), Mediterranean (Italy, France)

TABLE 5. Summary of literature data of cnidomes described for “ E. crocea ” (white rows) and “ E. ralphi " (shaded rows). The nematocysts of E. crocea by Schuchert 2010 (p. 360) have the same dimensions of those by Schuchert (1996) and Millard (1975). Measurements in micrometers, min – max. (*): only length reported.

Author (Original reference to species)	Heterotrichous anisorhiza	Basitrichous isorhiza	Stenotele Stenotele (large) (small)	Desmoneme	Microbasic eurytele
Ewer (1959) length ( Tubularia warreni ) width	9.5 *	9.0 8.0	8.0 7.0 (stenotele)	5.0 4.0
Brinckmann-Voss (1970) length ( Tubularia crocea ) width			present	present

TABLE 6. Measurements of nematocysts, in micrometers, by locality. Minimum value – maximum value (average value ± Standard deviation). Number of measurements for each cell of the table is 100, except for those marked with *. (* 1 n = 26, * 2 n = 2, * 3 n = 40, * 4 n = 20.)

	width	3.11–5.03 (4.03±0.32)	2.84–4.9 (3.77±0.43)	3.11–4.89 (3.94±0.3)	3.38–5.65 *1 (4.34±0.53)	3.16–6.04 (4.32±0.57)
Aboral	Desmonemes length	4.18–5.64 (4.85±0.3)	3.5–5.5 (4.59±0.43)	4.46–5.79 (5.1±0.33)	3.9–5.98 (4.85±0.38)	4.46–6.0 (5.19±0.29)
tentacles	width	2.62–3.84 (3.35±0.26)	2.41–4.11 (3.17±0.33)	2.84–4.48 (3.63±0.28)	2.69–4.62 (3.48±0.33)	2.94–4.54 (3.62±0.28)

TABLE 7. Summary of the genetic polymorphism observed for COI and ITS 1 + 5.8 S from the 22 samples of specimens of Ectopleura from the southwestern Atlantic Ocean in a total of five sampled localities.

	Length in bp	Haplotypes detected	Variable Sites	Mut1	Singl2	PIS3	Indel4	S/NS5	π/sd6
COI	489	11	20	20	8	12	0	15/5	0.012/0.001
ITS1+5.8S	332	3	1	1	1	0	3	-	0.001/0.0007

TABLE 8. Pairwise comparisons of localities for the CO 1 marker (10,000 permutations) of specimens of Ectopleura from the southwestern Atlantic Ocean showing no significant genetic differences between them (p> 0,05). Above the diagonal are the genetic distances expressed as corrected average of pairwise differences; below are the pairwise fixation indexes. Probability values are given in parentheses.

Macaé (RJ)	Juréia (SP)	Paranaguá (PR)	Bombas (SC)	Mar del Plata (ARG)
Macaé (RJ)	4.00 (0.10)	3.00 (1.00)	4.00 (0.10)	4.78 (0.10)
Juréia (SP) 0.76 (0.10)		5.00 (1.00)	6.00 (0.33)	7.00 (0.10)
Paranaguá (PR) 0.64 (0.99)	0.82 (0.99)		5.00 (0.33)	6.00 (0.25)
Bombas (SC) 0.76 (0.10)	0.86 (0.33)	0.82 (0.33)		7.00 (0.10)
Mar del Plata (ARG) 0.70 (0.10)	0.76 (0.10)	0.64 (0.10)	0.76 (0.10)