New Insight into the Genus Cladocroce (Porifera, Demospongiae) Based on Morphological and Molecular Data, with the Description of Two New Species

Abstract

During scientific expeditions in Indonesia and Vietnam, several sponge specimens belonging to the genus Cladocroce were collected. The integration of morphological and molecular analyses, incorporating species delimitation models (ABGD, ASAP, and bPTP) and phylogenetic approaches using three molecular markers (COI, 28S, and 18S–ITS1–5.8S–ITS2–28S), allowed us to discriminate three congeneric species. Two of these species (C. burapha and C. pansinii sp. nov.) were supported by morphological and molecular data, whereas a third species (C. lamellata sp. nov.) was delimited by morphological data only. We formally describe two new species, C. pansinii sp. nov. and C. lamellata sp. nov. C. aculeata is a newly recorded species for Indonesia and the first documented finding after the original description. The re-examination of the type material of C. burapha, and indirectly the molecular approach, allowed us to confirm that C. burapha lives in sympatry with C. pansinii sp. nov. in Vietnam and with C. lamellata in Indonesia. Thanks to these findings, we relocated the paratype of C. burapha to the new species described here, i.e., C. pansinii sp. nov.

1. Introduction

Genus Cladocroce Topsent, 1892, belongs to the family Chalinidae Gray, 1867 (ord. Haplosclerida Topsent, 1928), and its distinctive characteristic is the presence of multispicular ascending tracts within a subisotropic reticulation of oxeas [1]. The genus comprises species mostly typical of deep water [2]. At present, out of the 19 recognized species [3], 8 eight are found in tropical areas all around the world, and only five of these have been recorded in shallow waters [4]. Among these tropical shallow representatives, C. burapha Putchakarn, de Weerdt, Sonchaeng & van Soest, 2004, may be found in brackish environments with fluctuant periods of lower salinity such as coastal lagoons and mangroves [5], while the others are related to coral reefs and rocky substrates (e.g., [4,6,7]).

Surveys conducted in Ha Long Bay (Vietnam) (see [8,9]) and North Sulawesi (Indonesia) (see [5]) allowed us to collect several samples of Cladocroce, some of them ascribed to two new species, which are described here. Moreover, C. aculeata Pulitzer-Finali, 1982, is a new record for Indonesia and the first documented finding after the original description.

Integrative taxonomy approaches [10], based on DNA barcoding combined with morphological characterizations, were applied to shed light on the relationship between putative Cladocroce species, very closely aligned with C. burapha, living in sympatry in Vietnamese marine lake systems and Indonesian coastal reef and mangrove environments. Investigations based on morphological and molecular evidence, conducted on C. burapha (holotype and paratype) reference material, together with recent specimens from Hawai’i (USA) [7,11,12] and newly collected samples unveiled conspecificity discrepancies between C. burapha’s holotype and paratype with consequent proposed taxonomical changes.

2. Materials and Methods

2.1. Sponge Sample Collection

The specimens studied here were sampled from the Indonesian Archipelago and in Vietnam (Figure 1). The Indonesian samples were collected in the frame of a cooperation agreement between Sam Ratulangi University (Indonesia) and the Polytechnic University of Marche (Italy) at Bunaken National Park (North Sulawesi) (see [13] for site descriptions) in the period 2005–2008 and on Bangka Island (North Sulawesi) in 2011 (see [5]).

The Vietnamese samples were collected in Ha Long Bay (North Vietnam) along the coastal areas and in marine karst lakes during field campaigns performed in April 2003, September 2003, April 2004, and August 2018, as a cooperative affiliation between the Polytechnic University of Marche, the University of Genova, and the Institute of Marine Environment and Resources (IMER) for the study of biodiversity and conservation in coastal areas of Vietnam (see [8,9] for site descriptions). All samples were collected by scuba diving from a 0–40 m depth and were fixed in 70% or 100% ethanol.

2.2. Morphological Characterization

The spicule complement and skeletal architecture were examined under light and scanning electron microscope (SEM); spicule preparations and hand-cut sections of sponge portions were performed following the methodologies in Núñez-Pons et al. [14]. For each spicule type, measurements were obtained from 30 spicules and are reported as the range of the smallest length–(mean ± standard deviation)–largest length × smallest width–(mean ± standard deviation)–largest width. For SEM analyses, dissociated spicules were transferred onto stubs and sputter coated with gold. SEM preparations were observed under a Philips XL 20 (Polytechnic University of Marche, Ancona, Italy).

The type material was deposited at the Museo Civico di Storia Naturale “G. Doria” of Genova (MSNG).

The reference material used for comparison was obtained from the following institutions: BIMS-I Institute of Marine Science, Burapha University, Thailand; ZMA Zoological Museum Amsterdam, University of Amsterdam, The Netherlands.

2.3. Molecular Investigation

2.3.1. DNA Extraction, Amplification, and Sequencing

The genomic DNA of the sponge samples was purified using the Quick-gDNA Miniprep Kit (ZYMO RESEARCH, Irvine, CA, USA) following the manufacturer’s protocols and was stored at −20 °C. Three gene marker regions were targeted. The mitochondrial Cytochrome C Oxidase subunit 1 (COI) was amplified for the standard COI partition (~640 bp) with the universal Folmer [15] dgHCO2198 (5′-GGTCAACAAATCATAAAGAYATYGG-3′) and dgLCO1490 (5′-TAAACTTCAGGGTGACCAAARAAYCA-3′) primers and for the Erpenbeck’s ‘I3-M11’ extension (~560 bp), overlapping ~60 bp with Folmer’s 3′ partition, using the sponge-specific primers PorCOI2fwd (5′-AATATGNGGGCNCCNGGNATNAC-3′) and PorCOI2rev (5′-ACTGCCCCCATNGATAAAACAT-3′) [16]. The nuclear fragment covering 18S-ITS1-5.8S-ITS2-28S—hereinafter, referred to as ITSs (~850 bp)—was amplified with RA2 (5′-ACTGCCCCCATNGATAAAACAT-3′) priming on the 3′ terminus of the 18S small ribosomal subunit and the ITS2.2 primer (5′-CCTGGTTAGTTTCTTTTCCTCCGC-3′) targeting the 5′ terminus of the 28S large ribosomal subunit [17]. The 28S C region was amplified with the primers 28S-C2-fwd (5′-GAAAAGAACTTTGRARAGAGAGT-3′) and (28S-D2-revc 5′-TCCGTGTTT CAAGACGGG-3′), yielding a ~340 bp fragment [18]. PCR amplifications were performed in 25 μL reactions containing 2.5 μL (2 mM) of buffer, 1 μL dNTPs (2 mM), 0.8 μL (20 μg/μL) of BSA, 0.3 μL (5 U/μL) of Roche Taq DNA Polymerase, 0.8 μL (10 mM) of each primer, and 1 μL of template DNA, following a thermocycling profile of 5 min at 95 °C; followed by 35 cycles of 30 s at 94 °C, 45 s at 50–57 °C and 90 s at 72 °C; and a final extension for 10 min at 72 °C. The amplified PCR products were checked in 1% agarose gels under UV light and quantified on NanoDrop^® at Stazione Zoologica Anton Dohrn, Naples, Italy.

Sanger sequencing was performed bi-directionally at Servizio di Biologia Molecolare (RIMAR Dept, Stazione Zoologica Anton Dohrn, Naples, Italy.) on a Thermo Fisher Scientific 48 capillary ABI 3730xl DNA Analyzer, using 4.5 micromolar of each of the corresponding PCR primer pairs and 15 femtomoles/uL of the DNA template. Chromatograms were visualized, assembled, and corrected on Geneious Prime^® 2022.2.2 (http://www.geneious.com) [19], and poriferan origin was checked by BLAST against NCBI GenBank (http://www.ncbi.nlm.nih.gov/BLAST/ accessed on 23 October 2022) [20]. Sequences and specimen descriptions were deposited in GenBank (accession numbers: COI entries: OP950351–OP950360; 28S entries: OP955945–OP955951; and ITSs entries: OP955954–OP955959) and Sponge Barcoding Project databases for public access.

2.3.2. Phylogenetic Analysis

The resulting sequences from the Folmer and Erpenbeck partitions were concatenated to obtain a final COI fragment of ~1100 bp, which seems to afford better taxonomic resolution in Porifera and other diploblasts [21]. Reference sequences for the three markers, COI, 28S, and ITSs, from sponges in the Genus Cladocroce, along with the outgroup Haliclona (Gellius) cymaeformis (Esper, 1806) in the family Chalinidae were downloaded from GenBank (Table S1) and aligned against our sequences correspondingly for each marker with MAFFT [22] under default settings in Geneious Prime^® 2022.2.2 [19]. The alignments were manually trimmed, with COI sequences yielding 1047 base pairs (bp) long, 28S spanning 531 bp, and ITSs 815 bp. COI sequences were checked for stop codons using the translate function (DNA to protein) in Geneious Prime^® 2022.2.2 [19]. Alignments of the non-coding genes (28S and ITSs) were curated from blocks of gap positions’ variability on Gblocks 0.91b (http://phylogeny.lirmm.fr/phylo_cgi/one_task.cgi?task_type=gblocks accessed on 23 November 2022), to clean poorly aligned positions and eliminate divergent regions, applying stringent settings [23]. For the 28S, Gblocks did not result in an improvement of the data; hence, the original alignment (531 bp) was used for the phylogeny and species delimitation approaches below. Instead, for ITSs, the best output after Gblocks yielded an alignment of 505 bp, which was used for downstream analyses. All three definitive sequence sets were concatenated, resulting in a 2128 bp-length alignment (28S = 531 bp and COI = 1047 bp, ITSs = 505 bp).

Maximum likelihood (ML) trees were built for each marker separately and for the concatenated alignment (see below) with IQTree with 1500 ultrafast bootstrap replicates [24], under the most-accurate genetic models: TN + F for 28S, HKY + F for COI, and K2P + G4 for ITSs (Gblocks curated spanning 505 nt), according to ModelFinder [25]. Bayesian inference (BI) trees were constructed with MrBayes [26] with the GTR GAMMA model and Markov chain Monte Carlo (MCMC) set for 1,100,000 generations, 4 chains, a 200 generations’ sampling frequency, and 500 burn-in values (Supplementary File S2, Table S3). Phylogenetic analytic tools were run on CIPRES (http://www.phylo.org/ accessed on 25 November 2022) [27], and ML and BI trees were visualized and edited with Figtree (http://tree.bio.ed.ac.uk/software/figtree/ accessed on 25 November 2022). Since the representation of all markers was uneven across samples, the trees built with the concatenated data were not used for phylogenetic studies or species delimitation.

2.3.3. Species Delimitation

Species delimitation was calculated on COI, 28S, and curated ITSs data using three methods via web servers: Automatic Barcode Gap Discovery (ABGD), Assemble Species by Automatic Partitioning (ASAP), and Bayesian implementation of the Poisson-Tree-Processes model (bPTP).

ABGD was performed with the Kimura-2 parameter (K2P) and relative gap width (X = 1.5) (www.abi.snv.jussieu.fr/public/abgd/ accessed on 23 November 2022). ABGD calculates the barcode gap by setting a series of prior intraspecific divergences, which are iteratively refined [28].

ASAP was run with Kimura (K80) ts/tv set to 2 (https://bioinfo.mnhn.fr/abi/public/asap/asapweb.html accessed on 23 November 2022). ASAP computes species partitions based on pairwise genetic distances, together with the probability of panmixia (P-val), relative gap width (W), and ranked results by the ASAP score: the lower the score, the better the partitioning is [29]. The outcomes of the number of species predicted by the ASAP 1st and ASAP 2nd scores were selected.

bPTP was fed with ML trees constructed in IQTree and processed for 50 million generations, with a 100 generations’ sampling frequency and 25% burn-in (https://species.h-its.org/ptp/ accessed on 23 November 2022). Speciation events were inferred in PTP using MCMC based on a shift in the number of substitutions between internal nodes [30].

3. Results

Systematics:

• Class Demospongiae Sollas, 1885;
• Order Haplosclerida Topsent, 1928;
• Family Chalinidae Gray, 1867;
• Genus Cladocroce Topsent, 1892;
• Type species: Cladocroce fibrosa Topsent, 1892: 72;
• Cladocroce aculeata Pulitzer-Finali, 1982

(Figure 2).

Material examined: BUNA440, Bunaken, North Sulawesi, Indonesia, 05/2007, depth not stated.

Description: Small fragment of about 2 × 2 cm and 2–3 mm thick obtained from a convoluted, lamellate sponge (Figure 2A). Surface hispid with conules scattered on one side and numerous oscules on the opposite side. Consistency is firm, but elastic. The color is violet in life (Figure 2A) and white in alcohol.

Skeleton: The ectosomal skeleton is a regular, isotropic, tangential network, formed by unispicular meshes (Figure 2B,C). The choanosomal skeleton is formed by multispicular, anastomosing fibers (45–175 μm thick), and ascending, parallel fibers (Figure 2B,D), whose extremities create the conules (Figure 2B). Among the ascending fibers is the regular isotropic network (Figure 2B,D).

Spicules: Oxeas slightly curved or straight, with acerate, blunt, and stepped tips (Figure 2E). They measure 125–(154.3 ± 12.9)–180 × 2.5–(6.6 ± 2.5)–10 μm.

Distribution and ecology: The species was described as from the Great Barrier Reef (Lizard Island, Australia) and later collected from Papua New Guinea by Smith et al. [31]. In Indonesia, the species was associated with coral reefs.

Remarks: Morphological characterization: The study of the holotype (MSNG 46940) of C. aculeata, described by Pulitzer-Finali [32], confirmed the identification in terms of shape and size of the spicules (120–160 × 4–6.8 μm); also, the sizes of the fibers were comparable (50–170 μm in thickness). The sponge’s external morphology fits with the holotype as well; Pulitzer-Finali [32] described his small fragment as “a curved lamella […] apparently belonging to a tubular sponge”. Smith et al. [30] reported in this species the presence of molecules inhibiting human kinesin, but did not describe or illustrate the sponge. This is a newly recorded species for Indonesia and the first documented finding after the original description.

Cladocroce burapha Putchakarn, de Weerdt, Sonchaeng & van Soest, 2004

(Figure 3).

Material examined: MA6, Bangka Is., North Sulawesi, September 2011, 0–1 m depth. MA19 (a, e), Bangka Is., North Sulawesi, September 2011, 0–1 m depth. HL 116/18, Hang Trai Island—out Hang Du I Lake, Ha Long Bay, Vietnam, 22 August 2018. HL 119/18, Hang Trai Island—out Hang Dui I Lake, Ha Long Bay, Vietnam, 22 August 2018, 1 m depth. HL 121/18, Hang Trai Island—out Hang Dui I Lake, Ha Long Bay, Vietnam, 22 August 2018, 1 m depth. HL 139, Bo Hon Island—Bui Xam Lake, Ha Long Bay, Vietnam, 24 April 2004, 1 m depth. HL 139/18, out Bui Xam—Bo Hon Island, Ha Long Bay, Vietnam, 23 August 2018, 1 m depth.

Additional material: Holotype BIMS-I 1382.

Description: Cladocroce with massive, tubular (Figure 3A,C), or repent habit (Figure 3B). Oscules are round (3–8 mm in diameter) and raised. The sponge surface can be smooth or with shallow ridges and grooves (e.g., HL 139 in Figure 3A). In situ, the external color was light green to aquamarine. The inner color is generally light tan also in the preserved (both in ethanol and dried) specimens. Consistency is rather soft.

Skeleton: The ectosome is an isodictyal, tangential reticulation with scarce spongin. A very similar reticulation is present also in the choanosome, reinforced by a second, rather irregular reticulation of thick plurispicular tracts forming roundish or elongated meshes (Figure 3D). The plurispicular fibers range from 16.1 to 80 µm in thickness (

Table 1. Cladocroce burapha Putchakarn et al. 2004: morphological features of the studied specimens.

Specimen	Locality	Shape	Color	Oxea Size (µm)	Fiber Thickness (µm)	Mesh Diameter (µm)	Depth (m)	References
BIMS-I1382 Holotype	Gulf of Thailand	Tubulo-ramose	Off-white to light cream	105–117 × 6	23.5–59	131.4–499.6	2	[6]
BIMS-I1383	Gulf of Thailand	Tubulo-ramose	Light blue	87.5–110 × 6–9.5	/	/	low tide	[6]
MA 6	Bangka Is. North Sulawesi	Cylindrical, repent	Light green to aquamarine	90–(101 ± 3.7)–110 × 2.5–(3.9 ± 1.2)–5	30–75	/	0–1	Present paper
MA 19a	Bangka Is. North Sulawesi	Tubular, repent	Light green to aquamarine	95–(105.5 ± 5.5)–112 × 2.5–(4.2 ± 1)–5	22.7–46.4	207.4–562.9	0–1	Present paper
HL 116/18	Out Hang Du I lake, Vietnam	Tubular, repent	Light green to aquamarine	90–(102 ± 6.7)–115 × 3.7–(4.5 ± 0.6)–5	/	/	1	Present paper
HL 119/18	Out Hang Du I Lake, Vietnam	Tubular, repent	Light green to aquamarine	80–(97 ± 8.1)–115 × 3.7–(4.3 ± 0.6)–5	/	/	1	Present paper
HL 121/18	Out Hang Du I Lake, Vietnam	Tubular, repent	Light green to aquamarine	90–(100.8 ± 6.3)–110 × 2.5–(4.5 ± 0.8)–5	16.1–69.1	104–562.1	1	Present paper
HL 139	Bui Xam Lake Vietnam	Massive, tubular	Light green to aquamarine	80–(97.3 ± 8.5)–110 × 2.5–(3.9 ± 1)–5	20–80	/	1	Present paper
HL 139/18	Bui Xam Lake Vietnam	Tubular, repent	Light green to aquamarine	95–(108 ± 16.5)–160 × 5	26.1–58.6	98.7–177	1	Present paper

).

Spicules: Oxeas straight or slightly curved (Figure 3E) (80–160 × 2.5–9.5 µm). Points are generally sharp, but, in some specimens, blunt extremities are common, determining the presence of stylote and strongylote forms. All sample measurements of oxeas are in Table 1.

Distribution and ecology: The species seems to prefer shallow (maximum depth 15 m) and calm (bays, marine lakes, mangroves) waters, with variable salinity. Specimens from Indonesia were collected in mangrove forests, while those from Vietnam in salt-water lakes and shallow reefs [5,9]. C. burapha was described from the eastern coast of the Gulf of Thailand and recorded also in Vietnam and Indonesia. In 2017, Núñez-Pons et al. reported this species in Hawai’i, but this last identification is rejected here based on new data (see below) [7].

Remarks: Morphological characterization: The examination of the holotype material (BIMS-I1382) allowed us to confirm the specific determination. Specimens from Vietnam and Indonesia reported here fit with the holotype of C. burapha in the general shape and color and in the skeletal elements: the shape and size of the oxeas are similar and also the skeletal organization with quite thin fibers and quite small mesh diameter (Table 1). The taxonomic status of the other specimen (BIMS-I1383) described in Putchakarn et al. [6] should be maintained as C. burapha.

On the contrary, on the basis of the present findings, we do not agree with considering the paratype of C. burapha (ZMA Por. 17921) conspecific with the holotype; the details are clarified below, in the C. pansinii sp. nov. remarks.

Cladocroce lamellata Bertolino & Calcinai sp. nov.

urn:lsid:zoobank.org:act:C51F2F4E-C125-4633-9B60-F26AB4849545

(Figure 4).

Material examined: Holotype: MSNG 61503, TH A, Thangiung Husi, Bangka Island (North-Sulawesi), 10 October 2011 about 20 m depth.

Paratype: MSNG 61504, TH B, Thangiung Husi, Bangka Island (North-Sulawesi), 10 October 2011 about 20 m depth.

Other material: TH C, Thangiung Husi, Bangka Island (North-Sulawesi), 10 October 2011; TH D, Thangiung Husi, Bangka Island (North-Sulawesi) 9 October 2011 about 20 m depth; BA 7, Bangka Island (North-Sulawesi), 12 September 2011, 20 m depth; PH 9, Timur, Bunaken Island, (North-Sulawesi), 13 January 2005, 20 m depth.

Diagnosis: Cladocroce characterized by lamellate shape and with peculiar arrangement of the openings of the aquiferous system, separated on the two sponge sides. The skeleton is characterized by large primary fibers of the choanosome up to 500 µm.

Description: Lamellate sponge (Figure 4A), anchored directly to the substrate; it is violet in situ and in the dried state. Only one sample is yellowish (TH C). Alcohol-preserved samples are light cream colored. The holotype consists of two portions of about 8 × 4 and 5 × 4 cm. The paratype consists of a fragment of about 8 × 8 cm. The sponge is hard and firm; the general appearance of the sponge is different on the two sides. The inhalant side has a shaggy surface, with numerous scattered conules 2–3 mm high. A thin ectosomal membrane covers the sponge surface; ostia are not clearly visible. Conules are formed by choanosomal tracts that run to the surface, making it hispid (Figure 4C). The exhalant side is less shaggy, with shorter conules (<1 mm), and bears numerous oscula (1.5–2 mm in diameter) (Figure 4E).

Skeleton: The ectosomal skeleton is an isodictyal, tangential reticulation (Figure 4B,C,E) supported by multispicular (about 20 spicules) tracts, 40–130 µm thick, which protruding from the surface and create the conules (Figure 4C); between the multispicular tracts an unispicular, a less regular skeleton is present (Figure 4D,F). The deeper tracts are thicker, up to 500 µm in diameter (Figure 4F), and create polygonal meshes of 400–(977.5 ± 524)–1900 µm.

Spicules: Oxeas slightly curved, with acerate and hastate tips (Figure 4G,H) (125–195 × 2.5–7.5 µm). All sample measurements of oxeas are in

Table 2. Spicules measurements of Cladocroce lamellata sp. nov.

Samples	Oxeas (µm)
Holotype (MSNG 61503) TH A	135–(147.5 ± 9.2)–195 × 3.7–(4.5 ± 0.6)–5
Paratype (MSNG 61504) TH B	125–(148.2 ± 9.7)–165 × 2.5–(5.3 ± 1.7)–7.5
BA 7	145–(150.2 ± 47)–160 × 2.5–(4.1 ± 0.8)–5
PH 9	155–(162 ± 8.3)–190 × 3.7–(4.7 ± 0.5)–5
TH C	130–(150 ± 13.3)–175 × 2.5–(3.8 ± 1)–5
TH D	145–(154.5 ± 5.5)–165 × 3.7–(5 ± 1)–7.5

.

Etymology: The new species is named after its characteristic lamellate shape.

Remarks: There are only five valid species from the Indo-Pacific Ocean [3]: C. aculeata Pulitzer-Finali, 1982, C. tubulosa Pulitzer-Finali, 1993, C. burapha Putchakarn et al., 2004, C. reina Aguillar-Camacho & Carballo, 2010, and C. incurvata Lévi & Lévi, 1983 (

Table 3. Cladocroce genus: morphological features and distribution of hitherto known species.

Specimen	Locality	Shape	Color	Oxea Size (µm)	Fiber Thickness (µm)	Mesh Diameter (µm)	Depth (m)	References
C. aculeata Pulitzer-Finali, 1982	Indonesia	Lamellate	Violet	120–180 × 2.5–10	45–175	/	Shallow water	Present paper, [32]
C. attu Lehnert & Stone, 2013	Alaska	Funnel	Golden brown	158–183 × 14–16	/	/	358	[4]
C. burapha Putchakarn et al., 2004	Thailand, Indonesia, Vietnam	Tubulo-ramose, massive, tubular, repent	Off-white to light cream, light blue, light green to aquamarine	80–160 × 2.5–9.5	16.1–80	98.7–562.9	low tide, 0–2	[6], present paper
C. caelum Santo et al., 2014	Brazil	Massive tubular	Blue	62–86 × 2–5	24–80	/	1	[4]
C. fibrosa (Topsent, 1890)	Azores	Vase shaped	Grey	600 × 18	/	/	1300	[1]
C. gaussiana (Hentschel, 1914)	Antarctic	Tubular	Yellow	Oxeas: 230–285 Toxas: 80–180	/	/	200	[1]
C. guyanensis Van Soest, 2017	Guyana Shelf	Flabellate	Pale orange-brown	336–414 × 9–19	80–120	/	104–130	[33]
C. hyaline (Lundbeck, 1902)	Faroe Plateau	Erect, lamelliform	Hyaline	268–315 × 10–14	/	/	471	[34]
C. incurvata Lévi & Lévi, 1983	New Caledonia	Lamellate	Ochre	180–200 × 8–10	/	/	170–190	[1]
C. infundibulum Lehnert & Stone, 2013	Alaska	Funnel	Golden brown to light brown	232–281 × 19–23	42–128	/	180	[35]
C. kiska Lehnert & Stone, 2013	Alaska	Flabellate	Golden brown	Oxeas: 252–343 × 10–18 Sigmas: 17–25	/	/	235	[4]
C. lamellata sp. nov.	Indonesia	Lamellate	Violet, yellowish	125–195 × 2.5–7.5	40–130 up to 500	400–1900	20	Present paper
C. osculosa Topsent, 1927	Portugal	Lamellate	Brown	225 × 9	/	/	749–310	[1]
C. pansinii sp. nov.	Vietnam, Thailand, Hawai’i	Massive tubular, tubulo-ramose, repent	Light green, light blue, light grey, violet	125–200 × 3.7–15	25–376	166–991	1–15	Present paper, [6,7]
C. parenchyma (Lundbeck, 1902)	Greenland	Tubular	Yellow	239 × 9–12	/	/	2260	[1]
C. reina Aguilar-Camacho & Carballo, 2010	Mexico	Cushion, encrusting	Green, sky blue	130–175 × 5–7.5	30–150	150–350	3	[1]
C. spathiformis Topsent, 1904	Azores	Tubular	Transparent	375 × 17	/	/	1165	[1]
C. spatula (Lundbeck, 1902)	Greenland	Lamellate	Yellow	190–220 × 10–12	/	/	100	[1]
C. toxifera Lehnert & Stone, 2016	Alaska	Lobate	Reddish light brown	Oxeas: 241–297 × 12–23 Toxas: 36–139 × 3–7	50–230	/	93	[35]
C. tubulosa Pulitzer-Finali, 1993	Kenya	Tubular	Yellow	64–74 × 3.5	/	/	16	[1]
C. ventilabrum (Fristedt, 1887)	Behring Sea	Lamellate	Brownish	250	/	/	512	[1]

). This last species, described from New Caledonia, is a laminar sponge with a thin stalk, and it was found at a 400 m depth. Furthermore, the oxeas are different in size (180–220 × 8–10 µm), and the multispicular fibers reach 300 µm. Among the other Indo-Pacific species, typical of shallow waters, C. aculeata differs from this new species due to its smaller oxeas (120–160 × 4–6.8 µm, [32]), with rounded strongylote tips (Figure 2E). The analysis of the holotype confirmed the differences between C. aculeata and C. lamellata sp. nov. in the shape of the oxeas (constantly similar to strongyles in C. aculeata) and also in the skeletal organization considering that, in C. aculeata, the multispicular tracts of the main skeleton are thinner (50–170 μm). C. burapha (considered as indicated previously in the remarks of the species) differs from C. lamellata sp. nov. due to the growth form and color (Table 1). The spicular tracts are thinner than those of C. lamellata sp. nov. Furthermore, the sponge surface is different because C. burapha is smooth, without conules. C. tubulosa differs by its tubular shape and due to the smaller spicule size (67–74 × 3.5 µm). C. reina is a cushion-shaped sponge, recorded in the Tropical East Pacific Province; it differs also by its smooth surface, raised round oscula of 3–9 mm in diameter and also for the thinner multispicular tracts (up to 150 µm). The differences between C. lamellata sp. nov. and the other new species of the genus Cladocroce described in this paper are discussed in the remarks below.

Cladocroce pansinii Bertolino & Calcinai sp. nov.

urn:lsid:zoobank.org:act:BF872778-3692-4C8C-874B-06F42D719280

(Figure 5)

Holotype: MSNG 52833, HL 19, Dau Be Island—Coastal Site II, Ha Long Bay, Vietnam, 4 April 2003, 2–4 m depth, reef.

Paratype: MSNG 52834, HL 65, Cong Do Island—Hang Toi Dark Cave, Ha Long Bay, Vietnam, 14 September 2003, 1–1.5 m depth, semi-dark tunnel, rock.

Examined material: HL 60 and HL 70bis, Cong Do Island—Hang Toi Dark Cave, Ha Long Bay, Vietnam, 14 September 2003, 1–1.5 m depth. HL 95, Bo Hon Island—Bui Xam Lake, Ha Long Bay, Vietnam, 15 September 2003, 1.5 m depth, rock and mud. HL 110, Bo Hon Island—Hang Luong Lake (cove), Ha Long Bay, Vietnam, 16 September 2003, 2 m depth, rock and algae. HL 117/18, Hang Trai Island—out Hang Dui I Lake, Ha Long Bay, Vietnam, 22 August 2018, 2–3 m depth, rock and mud. HL 125/18, Bo Hon Island—Bui Xam Lake, Ha Long Bay, Vietnam, 23 August 2018, 2 m depth, rock and mud. HL 162/18, Me Cung Lake, Ha Long Bay, Vietnam, 24 August 2018, 2 m depth.

Additional material: Paratype of C. burapha ZMA POR 17921.

Diagnosis: Massive, with tubular processes. Tracts of oxeas of the choanosome up to 300 µm in thickness and wide meshes up to 991 µm. Oxeas 125–200 µm long and up to 12 µm in thickness.

Description: The sponge is generally massive, with a thick base from which stout digitations, with apical oscules, arise (Figure 5A,C). The oscules (2 to 8 mm in diameter) are at the digitation extremities, but are also scattered on the sponge body. In the latter case, they are bordered by thin rims up to 5 mm high. The holotype (HL 19) (Figure 5A) is massive with tubular processes, which often coalesce; these processes are up to 10 cm high and 3 cm across, tapering with conical tips (Figure 5A). In more sheltered conditions, as those occurring frequently in the marine lakes of Ha Long Bay, the sponge shows a repent habit (Specimen HL 95), producing also numerous, thin (2–3 mm in diameter), and long (15–20 cm) anastomosing processes (Figure 5B). The color in vivo of the Vietnamese specimens is light green and does not show much variation. Alcohol-preserved and dried specimens become beige to light brown. The consistency is firm and scarcely elastic.

Skeleton: The ectosome is an isodictyal network of triangular and quadrangular meshes with very few sparse spicules (Figure 5D,E). It is easily detachable in the dried specimens. In the choanosome, in proximity to the sponge surface, tracts of spicules organize to form a network of rounded or polygonal meshes, supporting the ectosome (see

Table 4. Cladocroce pansinii sp. nov.: morphological features of the studied specimens.

Samples	Locality	Shape	Color	Oxeas (um)	Fiber Thickness (µm)	Mesh Diameter (µm)	Depth (m)	References
Holotype (MSNG 52833) HL 19	Coastal Site II, Vietnam	Massive tubular	Light green	155–(174.4 ± 8.4)–185 × 7.5–(9.7 ± 1.6)–12.5	72.5–214	259.1–859.9	2–4	Present paper
Paratype (MSNG 52834) HL 65	Hang Toi Dark Cave, Vietnam	Massive tubular	Light green	155–(176.9 ± 9.4)–190 × 7.5–(10 ± 1)–12.5	68–289	255.6–987	1–1.5	Present paper
ZMA Por. 17921 Paratype	Gulf of Thailand	Tubulo-ramose	Light blue	141–171 × 6–7.5	25–100	252–991	15	[6]
SPO 29	Hawai’i	Tubular	Light blue	130–(147 ± 10)–167 × 5–(6.5 ± 0.9)–8.7	60–300	200–980	3–5	[7]
SPO 32	Hawai’i	Massive	Light grey	125–(138.2 ± 10.1)–157.5 × 3.7–(6.2 ± 1.5)–8.7	60–300	200–980	3–5	[7]
UF 3747	Hawai’i	Massive	Violet	138–147	/	/	/	[11]
HL 60	Hang Toi Dark Cave, Vietnam	Massive tubular	Light green	160–(174 ± 8.2)–200 × 7.5–(9.5 ± 0.8)–10	85.3–179.8	328.9–629.3	1–1.5	Present paper
HL 70 bis	Hang Toi Dark Cave, Vietnam	Massive tubular	Light green	150–(166.6 ± 8.5)–180 × 7.5–(9.7 ± 1.2)–12.5	55.1–94.5	208.6–771.1	1–1.5	Present paper
HL 95	Bui Xam Lake, Vietnam	Repent habit	Light grey	160–(177.8 ± 8.8)–200 × 7.5–(9.7 ± 1.2)–12.5	41–376	234–759	1.5	Present paper
HL 110	Hang Luong Lake (cove), Vietnam	Massive tubular	Light green	155–(166.3 ± 7.2)–175 × 5–(9.6 ± 2.6)–12.5	70–205	182.9–856.7	2	Present paper
HL 117/18	out Hang Dui I Lake, Vietnam	Massive tubular	Light green	145–(156.8 ± 8.4)–170 × 5–(8.4 ± 2.6)–12	/	/	2–3	Present paper
HL 125/18	Bui Xam Lake, Vietnam	Massive tubular	Light green	135–(155.8 ± 12.5)–180 × 7.5–(11.7 ± 2.8)–15	49–200	166–848	2	Present paper
HL 162/18	Me Cung Lake, Vietnam	Massive tubular	Light green	150–(167.6 ± 8.8)–185 × 7.5–(9.2 ± 1.1)–10	/	/	2	Present paper

for measurements) (Figure 5D,E). Deeper, the skeleton is formed by plurispicular tracts, which branch and anastomose without forming a regular reticulum (Figure 5F). The unispicular sub-isotropic reticulum is present also in the choanosome, between the plurispicular tracts (Figure 5F).

Spicules: Straight or slightly curved oxeas, generally with pointed ends (Figure 5G) (125–200 × 3.7–15 µm). One or both of the extremities may be blunted, and malformations are seldom observed. All sample measurements of oxeas are in Table 4.

Ecology: The species is rather common in Ha Long Bay. It was collected at a 2–4 m depth in very impoverished reef environments.

Etymology: Named after Prof. Maurizio Pansini in consideration of his fundamental contribution to sponge taxonomy.

Remarks: Morphological characterization: The other Indo-Pacific species of Cladocroce hitherto described [3] are: C. aculeata Pulitzer-Finali, 1982, C. incurvata Lévi & Lévi, 1983, C. tubulosa Pulitzer-Finali, 1993, C. burapha (Putchakarn et al., 2004), and C. reina Aguilar-Camacho & Carballo, 2010 (see previous remarks and Table 3).

Cladocroce aculeata, from the Great Barrier Reef, differs from the present species especially in the external morphology, due to the spiny processes reminded by the specific name, and in the spicules, which are much smaller (125–180 × 2.5–10 μm) and almost always with blunt extremities. C. incurvata differs from the present species as it is a lamellate sponge, ochre in color; it was described in New Caledonia from deep waters (170–190 m). C. tubulosa, from east Africa, has smaller spicule tracts (8–12 µm) and shorter spicules (not exceeding 75 µm in Pulitzer-Finali’s preparations) with respect to the new species. The other two species, C. burapha and C. reina, are close to C. pansinii sp. nov.

Cladocroce reina has been described from the Pacific coast of Mexico. It is a cushion-shaped or encrusting sponge, green or blue. Its oxeas are smaller in size (130–175 × 5–7.5 μm), and the choanosomal tracts are thinner (30–150 µm).

The main differences between the new species and C. burapha (as previously pointed out) are in the size of the oxeas (in C. burapha, the largest size of the oxeas does not reach the minimum value of the oxeas of C. pansinii sp. nov.) and in the diameter of the spicule tracts and of the meshes, which are always bigger in C. pansinii sp. nov. (see Table 1 and Table 4 for comparison).

The study of the type material of C. burapha supported us in defining the new taxonomic status of the specimen indicated as a paratype of C. burapha (ZMA Por. 17921) by Putchakarn et al. [6], assigning it to the species C. pansinii sp. nov. described here. Already, Putchakarn et al. [6] underlined the great variability between the holotype and paratype of C. burapha, especially in the size of the oxeas and in the coloration. Nevertheless, they found that their specimens were similar “with respect to form, consistency, skeletal architecture and shape of the oxeas” adding that “additional specimens may give more insight in the variability in different populations of C. burapha n. sp.” The re-examination of the type material confirmed the differences in the spicule size, but also in the size of the fibers and of the meshes (Table 1 and Table 4) (not discussed by Putchakarn et al. [6]), thus supporting the attribution of the paratype (ZMA Por. 17921) to C. pansinii sp. nov.

Núñez-Pons et al. [7] reported two specimens identified as C. burapha in Hawai’i; these sponges are described as light blue and grey, with oxeas, fibers, and meshes comparable in size to C. pansinii sp. nov. (Table 4). Based on the statements above, also these specimens should be identified as C. pansinii sp. nov.

C. pansinii sp. nov. differs from C. lamellata sp. nov. in the general shape, respectively massive and fan-shaped, and in the surface, which is smooth in C. pansinii sp. nov. and conulose in C. lamellata sp. nov. The spicules tracts and meshes are larger and particularly ticker in C. lamellata sp. nov. (see Table 3 and species descriptions).

Molecular Species Delimitation

We were able to gather molecular data from three of the four Cladocroce species reported here. Ten COI sequences (covering the Folmer and Erpenbeck partitions) in total were obtained from C. burapha (HL 116/18, MA 19a, HL 119/18, HL 121/18, and HL 139/18), C. lamellata sp. nov. (TH A, TH C, and TH D), and C. pansinii sp. nov. (HL 117/18 and HL 125/18), whereas, for the nuclear markers, six 28S barcodes were retrieved from the species C. burapha (116 HL/18, HL 119/18, and HL 139/18) and C. pansinii sp. nov. (117 HL/18, HL 125/18, and HL 162/18), while the ITSs yielded seven sequences, four from C. burapha (116/18, HL 119/18, HL 121/18, and HL 139/18) and three C. pansinii sp. nov. (HL 117/18, HL 125/18, and HL 162/18). The nuclear gene markers in C. lamellata were unsuccessful in yielding barcodes, and Sample TH B failed at any genetic amplification.

The COI fragments including the Folmer and Erpenbeck partitions 100% matched each other among the Vietnamese (HL 116/18, HL 119/18, HL 121/18, and HL 139/18) and Indonesian (MA 19a) C. burapha samples, saving a nucleotide disparity in position 876 differing 116 HL/18 and HL 139/18 from HL 119/18, HL 121/18, and MA 19a. These sequences revealed 93.696% similarity, displaying 66 variable sites with respect to the specimen SPO32 from Hawai’i (named C. burapha in [7]), C. lamellata sp. nov. (TH A, TH C, TH D) from Indonesia, and C. pansinii sp. nov. (HL 117/18 and HL 125/18) from Vietnam and 67 nucleotide divergencies with the specimen SPO29 (named C. burapha in [7]) and Cladocroce sp. UF_3747, UF_4051 and KBOA061118110 from Hawai’i [11,12]. C. pansinii sp. nov. Samples HL 117/18 and HL 125/18 yielded identical (100% similarity) sequences and 100% matched SPO 32 from Hawai’i (named C. burapha in [7]) and C. lamellata sp. nov. (TH A, TH C, TH D) from Indonesia, and they were different in one nucleotide (at position 885), hence 99.969% matching with SPO 29 (named C. burapha in [7]) and UF_3747, UF_4051, and KBOA061118110 from Hawai’i (named Cladocroce sp. in [11,12]) (Table S3). Cladocroce burapha Samples 116 HL/18, HL 119/18, and HL 139/18 had identical 28S sequences (531 bp). They differed in 127 nucleotide calls (75.049% similarity) with C. pansinii sp. nov. HL 117/18, HL 125/18, and Hl 162/18 from Vietnam and in 130 variable sites (74.656% similarity) from Cladocroce sp. UF_3747, UF_4051, and KBOA061118481 from Hawai’i [11,12]. 28S sequences in 117 HL/18, HL 125/18, and HL 162/18 (C. pansinii sp. nov.) 100% matched and revealed a difference of three nucleotide sites (99.399% similarity) from Cladocroce sp. UF_3747, UF_4051, and KBOA061118481 from Hawai’i [11,12] (Table S3). The ITSs fragments of Vietnamese C. burapha HL 116/18, HL 119/18, HL 121/18, and HL 139/18 were 100% identical. In the curated alignment (after Gblocks), the similarity of these ITSs sequences with respect to those of C. pansinii sp. nov. HL 117/18, HL 125/18, and HL 162/18 was between 75.273 and 75.636% (with 134, 135, and 136 variable sites, respectively) and differed in 127 nucleotide calls (76.909% similarity) with that of SPO32 from Hawai’i (named C. burapha in [7]). They differed by 36–37 variable sites (93.455–93.273% similarity) from SPO 32 from Hawai’i (named C. burapha in [7]) and were dissimilar in 134–136 nucleotide calls (75.636–75.273% similarity) from the ITSs sequences of C. burapha HL 116/18, HL 119/18, HL 121/18, and HL 139/18. Among the C. pansinii samples, HL 117/18’s ITS sequence had one nucleotide difference with HL 162/18 and three variable sites with HL 125/18, and these two last samples were different by two nucleotide calls, all three being 99.866–99.455% identical (Table S3).

The ABGD and ASAP models yielded similar outcomes of species delimitation. The 15 query COI sequences were divided into two groups, with HL 116/18, HL 139/18, HL 119/18, HL 121/18, and MA 19 clustered into Group 1; specimens HL 117/18, HL 125/18, SPO 32, KBOA061118110, TH A, TH C, TH D, SPO 29, UF 3747, and UF 4051 clustered into Group 2 (ABGD: P = 3.59 × 10⁻², barcode gap distance = 0.021; ASAP: P-val (rank) = 1.00 × 10⁻⁵; threshold dist = 0.021). In the 28S data, two partitions are proposed, Group 1 formed by HL 116/18, HL 119/18, and HL 139/18 and Group 2 comprising HL 117/18, HL 162/18, HL 125/18, KBOA061118481, UF 3747, and UF 4051 (ABGD: P = 1.00 × 10⁻¹, barcode gap distance = 0.128; ASAP: P-val (rank) = 1.02 × 10⁻²; threshold dist = 0.126). Similarly, the ITS set yielded two partitions, Group 1 with HL 116/18, HL 121/18, HL 139/18, and HL 119/18 and Group 2 with SPO 32, HL 117/18, HL 162/18, and HL 125/18 (ABGD: P = 1.00 × 10⁻¹, barcode gap distance = 0.106; ASAP: P-val (rank) = 2.38 × 10⁻²; threshold dist = 0.105). In the three markers, Group 1 represents C. burapha specimens, while Group 2 includes C. pansinii sp. nov. from Vietnam and specimens of Cladocroce from Hawai’i (incorporating C. lamellata sp. nov. in the COI set) cluster (Figure 4 and Table S1; Supplementary File S1); bTPT also supported these two major groups for COI, 28S, and ITS data, yet in the ITSs, bTPT suggested a third tentative species within Group 2, separating SPO 32 from HL 117/18, HL 162/18, and HL 125/18. We believe this sub-partition is not taxonomically significant (see below; Figure 6, Supplementary File S2).

The phylogenetic trees used in the analyses were constructed for the three separate sets, COI, 28S, and ITSs. Concatenated trees were not used, as the coverage of sequence data was uneven across all markers in some samples (e.g., TH A, TH C, and TH D lacked two marker genes). This generated some background cluttered branching between certain close nodes –especially in the ML phylogenies, more likely related to different levels of sequence data availability, rather than to real phylogenetic dissimilarities (Supplementary File S2). In lieu, the corresponding ML and BI trees had practically identical topologies in the three separate datasets. All trees supported a clear separation of C. burapha representatives from the C. pansinii sp. nov. cluster, with the latter including those previously classified as C. burapha and Cladocroce sp. from Hawai’i (now redefined as C. pansinii sp. nov.) and the newly described Vietnamese samples. In the COI phylogeny, C. lamellata sp. nov. was englobed within the C. pansinii sp. nov. cluster, whereas the ITS tree showed an internal branching within C. pansinii sp. Nov. separating SPO 32 from Hawai’i from the Vietnamese samples. This branching is unlikely to represent a separate species, but probably indicates local intraspecific variability (Figure 6, Supplementary File S2).

Based on nucleotide dissimilarities, genetic distances, ML and BI phylogenies, and ABGD, ASAP, and bPTP species delimitation models, the three marker genes clearly discriminated the C. pansinii sp. nov. cluster formed by Vietnamese Samples HL 117/18, HL 125/18, and HL 162/18 and Hawai’ian entries SPO 29, SPO 32, UF 3747, UF 4051, KBOA061118481, and KBOA061118110 from C. burapha Samples HL 116/18, HL 119/18, HL 121/18, and HL 139/18 from Vietnam and MA 19a from Indonesia. All the analyses also implied that C. burapha from Hawai’i (as in [7]; i.e., SPO 29 and SPO 32) should be redefined as C. pansinii sp. nov. Moreover, this is likely to be extended to several Cladocroce sp. Hawai’ian entries present in GenBank, i.e., UF 3747, UF 4051, KBOA061118481, and KBOA06111811 (as in [11,12]; discussed below).

According to the only genetic data available for C. lamellata sp. nov., the COI, there was no evident separation of C. lamellata sp. nov. (Specimens TH A, TH C, and TH D from Indonesia) from the cluster formed by C. pansinii sp. nov. composed of specimens from Vietnam (HL 117/18, HL 125/18), from Hawai’i (SPO 29 and SPO 32 named C. burapha in [7], and UF 3747, UF 4051, and KBOA061118110, named by [11,12] as Cladocroce sp.). This outcome, even if incongruent with the morphological data (see above), might represent only an inconclusive result, taking into consideration the often poor resolution of the COI marker in Porifera [36]. Indeed, we still support C. lamellata sp. nov. as a novel separate species, based on its divergent morphotype and diagnostic traits. Further considerations are addressed in the following sections.

4. Discussion

The coastal areas of Ha Long Bay (Vietnam) and North Sulawesi (Indonesia) comprise a wide assortment of environments, including shallow salty to brackish lakes, lagoons and mangroves, and deeper coral reefs, rocky bottoms, sandy beds, and plant or algal meadows [37,38]. Such heterogeneous habitat availability can be a driver of new species radiation, forcing organisms to acclimatize and diversify across a variety of ecological niches [39]. In this sense, the phylum Porifera has been demonstrated to be particularly rich and diverse in these areas and able to colonize most of the mentioned habitats, revealing a growing number of new species described in recent years [9,40]. Adding to these records, here, we provide the description of two new species in the genus Cladocroce, recognized through integrated taxonomy with morphological and molecular approaches.

C. pansinii sp. nov. was recognized as a new species, clearly differentiated from its closest sympatric congeneric, C. burapha, via morphological discriminant characteristics (larger size of the oxeas and larger spicule tracts and meshes) and through phylogenetic analyses and species delimitation.

The re-examination of the type material of C. burapha evidenced strong inconsistencies in the morphological diagnostic traits between the holotype and the paratype; instead, a close affinity of the paratype (ZMA Por. 17921) with C. pansinii sp. nov. (e.g., oxeas, fibers, and meshes comparable in size) is evident. The holotype (BIMS-I1382), in lieu, keeps strong closeness with specimens from Vietnam and Indonesia, which we identified as C. burapha. This fact suggests the assignment of the paratype of C. burapha to this new species, C. pansinii sp. nov. This conclusion is indirectly supported by the phylogenetic trees that clearly separate our specimens of C. burapha and C. pansinii into two distinct clusters.

Furthermore, specimens previously classified as C. burapha from Hawai’i (i.e., SPO 29, SPO 32; [7]) also should be redefined as C. pansinii sp. nov., as they display closer morphological affinity with C. pansinii sp. nov. and cluster together in the same phylogenetic branching and species delimitations. Molecular analyses also implied that Cladocroce sp. reported from Hawai’i as vouchers UF 3747, UF 4051, KBOA061118481, and KBOA06111811 from [11,12] take part of the same cluster above and should, hence, be defined as C. pansinii sp. nov. These samples again revealed larger oxeas, fibers, and mesh sizes than those found in C. pansinii sp. nov. ([7]; J. Vicente, personal communication).

This would make us aware of the possible incongruences between holotypes and paratypes. Because these inconsistencies, which often derive from vague descriptions on damaged material and/or assumed morphological variation/plasticity, can lead to an underestimation of biodiversity [41].

The other new species described here, C. lamellata sp. nov., was clearly separate from C. pansinii sp. nov. by morphological diagnosis (e.g., different shape, surface ornamentation, and size of the spicule tracts and meshes). Molecular data, based only on the COI marker, was not able to discriminate between these two species. The low molecular resolution of COI in Porifera, due to slow evolution rates in basal Metazoa [36], may not be sufficient to disclose delimitation between C. lamellata sp. nov. and C. pansinii sp. nov., which may be more closely related to each other than C. burapha. Nonetheless, other markers with faster mutation times, such as 28S and ITS, may have better resolution [42], but unfortunately, we were unable to amplify these fragments. Discordance between morphological and molecular species boundaries within the same genus has been already reported in sponges [43]. The ITS markers, for instance, have higher evolutionary rates, but may yield too much intraspecific intragenomic variability [44]. This could be the reason why certain methods based on phylogeny in our study (e.g., ML, BI, and bPTP) separated SPO 32 from the rest of C. pansinii sp. nov. This is unlikely to be a real species delimitation, but likely a local intraspecific polymorphism.

C. lamellata sp. nov. and C. pansinii sp. nov. have not been found at the moment living in sympatry with each other, but both instead do coexist with C. burapha. Considering their different ranges of distribution known at the present time, we may speculate C. lamellata sp. nov. could have evolved from the widely present C. pansinii sp. nov. in Indonesia. C. burapha, instead, seems to prefer shallow (maximum depth 15 m) and calm (bays, marine lakes, mangroves) waters and reveals smaller spicule sizes, with respect to C. pansinii sp. nov. and C. lamellata sp. nov., which can be found in deeper marine habitats and display larger spicules. Larger spicules in length and thickness have been correlated with populations living in deeper habitats with lower temperatures and higher silica concentration [45,46]; however, the depth difference in our sample set might not be as relevant to follow this tendency. Intraspecific variability has been generally usually considered as the main explanation in these cases of high phenotypic plasticity, but our results, coming from molecular analyses, highlight that speciation processes can be another way to interpret and disentangle these issues.

Our work expands the knowledge of species distribution along iconic hotspots of the Indo-Pacific Oceans and inspires research on marine biodiversity. Regarding the classification Porifera, our findings strongly supported the assumptions of other authors (e.g., [42]) for the lack of species resolution using one genetic marker and the necessity to combine morphological observations with a multilocus-based approach.