Shallow-water zoantharians (Cnidaria, Hexacorallia) from the Central Indo-Pacific

Abstract Despite the Central Indo-Pacific (CIP) and the Indonesian Archipelago being a well-known region of coral reef biodiversity, particularly in the ‘Coral Triangle’, little published information is available on its zoantharians (Cnidaria: Hexacorallia: Zoantharia). In order to provide a basis for future research on the Indo-Pacific zoantharian fauna and facilitate comparisons between more well-studied regions such as Japan and the Great Barrier Reef, this report deals with CIP zoantharian specimens in the Naturalis collection in Leiden, the Netherlands; 106 specimens were placed into 24 morpho-species and were supplemented with 88 in situ photographic records from Indonesia, the Philippines, and Papua New Guinea. At least nine morpho-species are likely to be undescribed species, indicating that the region needs more research in order to properly understand zoantharian diversity within the CIP. The Naturalis’ zoantharian specimens are listed by species, as well as all relevant collection information, and in situ images are provided to aid in future studies on zoantharians in the CIP.


Introduction
Zoantharians (Cnidaria: Anthozoa: Hexacorallia: Zoantharia) are a common component of benthos in subtropical and tropical coral reef systems, with many zooxanthellate species found in shallow waters of both the Atlantic and Indo-Pacific Oceans. Nevertheless, common understanding of zoantharian species diversity is relatively poor when compared to the hard corals (Scleractinia). This lack of knowledge is due to a variety of reasons, including (1) high levels of intraspecific morphological variation hindering reliable identification (Burnett et al. 1997, Reimer et al. 2004), (2) problems in performing histological examinations owing to sand being incorporated in the body walls of many zoantharian species , and (3) a confused taxonomic history as different researchers tried to properly classify and understand zoantharian diversity (Burnett et al. 1997, Reimer et al. 2004, Sinniger et al. 2005. Despite these problems, an understanding of zoantharian diversity and their corresponding taxonomy have slowly become clearer as molecular techniques have been implemented into zoantharian research. The first molecular works of Burnett and coworkers (Burnett et al. 1994(Burnett et al. , 1995(Burnett et al. , 1997 combined with the ecological and descriptive works of Ryland (Ryland andLancaster 2003, 2004) have led to more recent papers dealing with the molecular phylogeny of zoantharians (Reimer et al. 2004, 2012b, Sinniger et al. 2005, Swain 2010), resulting in a reassessment of zoantharian taxonomy (Fujii and Reimer 2011, Sinniger et al. 2013. Consequently, zoantharians are now perhaps the hexacorallian order for which the taxonomy most accurately reflects molecular phylogenetic understanding. However, whereas zoantharian supraspecific taxonomy and diversity is increasingly well understood, many problems remain at the species level (Reimer et al. 2007b), and total species diversity of zoantharians is still poorly known (Appeltans et al. 2012).
Recent work on zoantharians has focused on many regions of the Indo-Pacific, including Japan (Reimer 2010), Singapore (Reimer and Todd 2009), New Caledonia (Sinniger 2006), the Great Barrier Reef (Burnett et al. 1997, Reimer et al. 2011a and Palau (Reimer et al. 2014a). In the center between these regions lies the central Indo-Pacific "Coral Triangle" (Hoeksema 2007), including parts or all of Malaysia, Indonesia, Brunei, the Philippines, and Papua New Guinea, the Solomons, and Timor Leste. This region is believed to harbor the highest species diversity in hard corals of the order Scleractinia (Hoeksema 2007, Veron et al. 2009, and it is believed that other coral reef organisms likely have similar diversity patterns (Roberts et al. 2002). Despite this, shallow-water zoantharian species within the Coral Triangle have only briefly been reported on in scientific literature and only a few publications exist (e.g. Den Hartog 1997, Sinniger et al. 2005, Di Camillo et al. 2010, and most information is made up of photographs in aquarium handbooks (Fosså and Nilsen 1998) and field guides (Colin and Arneson 1995, Gosliner et al. 1996, Erhardt and Knop 2005. Therefore, efforts to compare the regional zoantharian fauna of the Indo-Pacific are hampered by this almost complete lack of published scientific distribution information. Basic data on zoantharians from the Coral Triangle, such as species lists and distribution records, are critical to achieve a comprehensive understanding of Indo-Pacific zoantharian diversity. The present study addresses this lack of Central Indo-Pacific (CIP) zoantharian data via examinations of specimen collections housed in Naturalis Biodiversity Center, Leiden, the Netherlands: RMNH (the former Rijksmuseum van Natuurlijke Historie) and ZMA (the former Zoologisch Museum van Amsterdam). These zoantharian collections are partly based on specimens from numerous surveys in Indonesia dating from the Snellius Expedition (1929)(1930) to a recent Marine Biodiversity Workshop in Lembeh Strait (2012), with the large majority of these specimens collected from coral reef environments. Despite the presence of these large and scientifically valuable collections, no previous effort has been made to comprehensively catalogue or examine these historical collections for over 80 years, which could also serve as base-line material for studies on biotic change . Here, for the first time, we report on the zoantharian specimens from Indonesia housed at Naturalis, and list shallow water species of the CIP, including specimen collection information. Our records are further enhanced by numerous in situ images from more recent fieldwork in Indonesia taken by the last author starting with the Snellius-II Expedition (1984)(1985). Finally, we discuss the shallow water zoantharian diversity of CIP in relation to information from surrounding regions, and make recommendations for future zoantharian research in the region.

Specimen collection
Zoantharian specimens from the Naturalis collections in Leiden (RMNH + ZMA) were collected primarily from expeditions to the Indonesia region, starting with the Snellius Expedition (1929)(1930). Our examinations showed 22 regions in which either specimens or photographic records were present. All specimen/record localities are shown in Figure 1 with location and reference details in Table 1.
Overview of field surveys from which order Zoantharia specimens examined in this study were collected.

Area
Year ( 2001 Hoeksema and Tuti (2001: 12-15) Includes southeast Bali and Lombok Strait. Coral biodiversity surveys in collaboration with RCO-LIPI and WWF Bali Indonesia Marine Program.

Specimen registration and identification
Examination of the registered (n=52) and unregistered zoantharian specimens (n=570) of the Naturalis collection showed that of a total 622 specimens, 105 were from Indonesia, with an additional four from the Philippines. Of these 109 specimens, 106 form the basis of this research, as we excluded three specimens that could not be conclusively identified as zoantharians. 88 photographic records of zoantharians specimens were also examined.
Although most species are from depths in the range of SCUBA (<40 m), we also included all Epizoanthus illoricatus Tischbierek, 1930 specimens, as although some specimens were from >40 m (and down to 190 m), the range of this species does extend into shallower (<40 m) depths. Additionally, three specimens of Parazoanthus collected by rectangular dredge from depths of 50-100 m were included in analyses. In this study, these 106 zoantharian specimens are collectively referred to as "shallowwater zoantharians".
All unregistered specimens were newly registered into the Naturalis collection in the course of our research. All specimens, newly registered or not, were re-identified by the first author. A list of specimens, their collection information, and Naturalis (RMNH Coel) registration numbers are given within each species' section. Descriptions of each species are given to aid in field and specimen identification, and are not formal taxonomic redescriptions.
Most zoantharian specimens were easily identifiable to genus level without microscopic examination. Species determinations were made consulting previous literature (listed with each species). However, many specimens were only identified to "confers with" (cf.) or "affinity" (aff.) levels. Asides from a few species (e.g. Palythoa heliodiscus), very few records of zoantharians had previously been formally reported from the CIP/ Coral Triangle region. Given these reasons, we followed recent research (Burnett et al. 1994, 1997, Reimer et al. 2006a, 2007b, Sinniger 2006, Sinniger et al. 2010, Reimer and Todd 2009, Reimer 2010, Reimer et al. 2011a, Fujii and Reimer 2011 from neighboring regions and used Zoanthus and Palythoa species names for which numerous references, molecular data and/or accurate descriptions were available, unless specimens and/or images clearly did not match with previously published information. Sizes of specimens are averages taken from measurements of 10 polyps per specimen, unless the specimen contained less than 10 polyps, in which case all non-damaged polyps were examined. For species' dimensions, average dimensions were taken from the overall average of specimens, unless there were less than three specimens within a species. In such cases, dimensions are stated only as a range (minimum to maximum).

Results
From specimen examination, the 106 Indonesian zoantharian specimens in the Naturalis collection supplemented with images were placed into 24 morphospecies, detailed below. Locations are in Indonesia unless otherwise noted, and all photographic images were taken by B.W. Hoeksema unless otherwise noted. Duplicate photographic images of the same species from the same site are counted as one record. Latitude and longitude are given when available.

Specimens and species
Abbreviations: NA=not available. Description. Non-incrusted zooxanthellate zoantharian that inhabits the outside of eunicid worm tubes (Haddon 1895), with a unique asexual form of "budding" (Ryland 1997). Easily recognizable as it is an epibiont on outside surface of eunicid worm tube, and has a reduced stoloniferous coenenchyme, long pale yellow-green or pale purple tentacles (n=approx. 40-50) with occasional fluorescent green markings and black tips, and light brown/purple to white ectoderm with similarly colored oral disks ( Figure 2A). Preserved specimens in this study had polyps of average 6.0 mm in height (range 2.5-14 mm), 3.2 mm in width (range 2-5 mm) (n=8 specimens examined [RMNH Coel 40361,40549,40550,40554,40556,40558,40566,40569], 10 polyps/specimen), and oral disks approximately 6 mm in diameter when expanded in situ (partially adapted from Reimer et al. 2011b). Specimen RMNH Coel 40566 had much larger polyps than other specimens (average height 10.6 mm, average width 4.3 mm), but this may be due to preservation in formalin as opposed to ethanol than to any phenotypic difference.
Previous records. This species has previously been reported from Zanzibar (type locality), Singapore (Reimer and Todd 2009), Taiwan (Reimer et al. 2011c(Reimer et al. , 2013a, Palau (Reimer et al. 2014a), southern Japan (Reimer et al. 2004, 2006a, Kamezaki et al. 2013, and is considered to have a very wide Indo-Pacific distribution. Remarks. Based on its wide Indo-Pacific distribution, it is very likely that this zooxanthellate species is much more common within the CIP than reported here. One possible reason for the lack of records from the CIP is that this species is most commonly found in the intertidal zone, which is under-sampled during SCUBA surveys. However, this species is also found to depths of 52 m (Kamezaki et al. 2013), although below the shallow littoral zone it rarely forms colonies >100 polyps.
Additionally, as preserved specimens of Zoanthus are notoriously hard to identify to species level, the large number of unidentified Zoanthus specimens in this study undoubtedly include some Z. sansibaricus colonies. This is also likely one important reason explaining the presence of comparatively more photographic records of this species in this study, as in situ identification of colonies with expanded oral polyps is easier than preserved specimen identification.

Distribution. Regions recorded in this study
Previous records. NA.
Remarks. This designation simply consists of all Zoanthus spp. specimens that could not be identified to species-level. It is likely this designation includes more than one species based on depths of specimens sampled. However, as preserved specimens were contracted (polyps closed) and many described Zoanthus spp. present no readily diagnostic external characters, identification to species level is not potentially possible without detailed molecular examination. Attempts at molecular identification also failed for these (and most other specimens), perhaps due to initial preservation in 10% seawater formalin for older specimens, or in ethanol with additives for newer specimens. Photographic records. NA. Description. Species in this genus are zooxanthellate, not incrusted, with a simple mesogleal sphincter muscle, and have non-erect, recumbent polyps that do not have lacunae or mesogleal canals, unlike Zoanthus species. Isaurus tuberculatus has tubercles on the exterior surface of polyps (=endodermal invagination) ( Figures 4A, B). For detailed discussion of I. tuberculatus, refer to Muirhead and Ryland (1985), with phylogenetic analyses in Reimer et al. (2008b).
Specimens examined in this study varied greatly in size from relatively large RMNH Coel 40567 (height 28-31 mm, width = 6-7 mm, n=2 polyps) to relatively small RMNH Coel 40473 (height average 10.6 mm, width average 2.9 mm, n=7 polyps). However, Isaurus polyps are known to vary greatly in size both between different colonies and within large colonies (Larson and Larson 1982;Muirhead and Ryland 1985;Reimer et al. 2008b). Furthermore, the two other valid Pacific Isaurus spp. asides from I. tuberculatus are both very distinct from these specimens, and found in Fiji and southwestern Australia, respectively. Thus, the identity of these specimens as I. tuberculatus is largely certain.
Distribution. Regions recorded in this study ( Figure 5). Moluccas (14), Lembeh Strait (17). Previous records. Originally described from the West Indies, this species is distributed throughout the subtropical and tropical Atlantic and Indo-Pacific (e.g. Muirhead and Ryland 1985), although populations in each ocean basin likely constitute different species (Reimer et al. 2008a). In the Indo-Pacific, it has previously been reported from the Great Barrier Reef, Fiji, Hawaii (summarized in Muirhead and Ryland 1985), and also from Indonesia (Sinniger et al. 2005), New Caledonia (Laboute and Richer de Forges 2004), and Japan (Reimer et al. 2008b).
Remarks. As seen in previous studies (Reimer et al. 2008b), it appears from the low numbers of specimens here that Isaurus is either somewhat rare throughout its range, or cryptic in nature (e.g. well-camouflaged), resulting in few reports of this species.  Description. Unique among zoantharians, species in this genus have an endodermal sphincter with brachycnemic mesentery arrangement. Polyps are only partially incrusted, with the oral end of polyps lacking incrustation ( Figures 4C, D). Phylogenetically, this genus is closely related to Isaurus (within family Zoanthidae). Zooxanthellate. Adapted from Herberts (1972), Reimer et al. (2012a).
Previous records. Species of this genus have been reported from Madagascar (Herberts 1972), southern Japan (Reimer et al. 2011a(Reimer et al. , 2012a(Reimer et al. , 2013b, and the southern Great Barrier Reef (Reimer et al. 2011a(Reimer et al. , 2012a. Remarks. This genus was originally described from Madagascar with the type species N. tulearensis Herberts, 1972. Subsequently, two species have been reported from Australia and Japan (Reimer et al. 2012a). As no specimens exist, it is impossible to determine if the Indonesian photographs constitute one or both of the species reported in Reimer et al. (2012a), or an as of yet undescribed species. Description. Originally described from the Torres Strait, Australia, this species was redescribed in detail in Ryland and Lancaster (2003).

Family
Although all specimens in this grouping match with previously reported P. mutuki based on sizes (average polyp height 9.6 mm, range 3-31 mm, average width 4.8 mm, range 2-8 mm, n=12 specimens) and overall morphology ('intermediae' or 'liberae' [Pax 1910]; visible capitulary ridges on closed polyps [Ryland and Lancaster 2003]) ( Figure 6B), we have identified all specimens in this study as "cf.". Recent work has shown the presence of more than two closely related species groups within P. mutuki (Reimer et al. 2006b(Reimer et al. , 2013a) that are exceedingly difficult to distinguish without molecular data. For this reason, we have preliminarily assigned "cf." to these specimens.
Previous records. Ryland and Lancaster (2003) in their treatment of P. mutuki also mentioned records from Fiji, and synonymized records of other species from Tuvalu (Gemmaria willeyi Hill & Whitelegge, 1898), eastern Australia (G. arenacea Wilsmore, 1909;P. yongei Carlgren, 1937;P. australiensis Carlgren, 1950) and Singapore (P. singaporensis Pax & Müller, 1956) with this species. However, asides from the specimens directly examined by Ryland and Lancaster, there is much confusion over the true identity of these species. For example, Ryland and Lancaster (2003) themselves state that G. willeyi is likely a Zoanthus species based on the figures in the original description. Ryland and Lancaster state "Probably only the use of genetic methods, so successfully applied by Burnett et al. (1997), will settle identities over wide geographic areas".
However, in the Pacific, records of this species with phylogenetic confirmation have previously been reported from the Great Barrier Reef in Australia (Burnett et al. 1997), Singapore (Reimer and Todd 2009), to the south Pacific coast of Japan (e.g. Reimer et al. 2006bReimer et al. , 2007b, New Caledonia (Sinniger 2006), and across to the Galapagos (Reimer and Hickman 2009), and thus it is known that this species has a very wide Indo-Pacific distribution.
Remarks. This species is likely common in Indonesia as in other regions such as Okinawa (Irei et al. 2011) and Taiwan (Reimer et al. 2011c). However, species delineation in Palythoa is confused due to the close phylogenetic relationships between P. mutuki, P. tuberculosa, and some other undescribed species, and a potential reticulate evolutionary history (Reimer et al. 2007b, Shiroma andReimer 2010, M. Mizuyama andJ.D. Reimer unpubl. data). Furthermore, distinguishing P. mutuki, from other, more distantly related species such as P. heliodiscus based solely on morphology is often difficult (Ryland and Lancaster 2003). For this study, we have included all "P. mutuki-like" specimens as one species group for convenience, although it is likely the specimens will encompass more than one species once the taxonomy of this genus is clarified. Photographic records. NA. Description. This group consists of two specimens that do not clearly fit with previously described Palythoa species. Both specimens have dimensions very different from other Palythoa specimens reported here; whether this is due to unusual fixation or relaxation methods, or to true phenotypic differences is unknown.
RMNH Coel 40508 ( Figure 6C) has very long 'liberae' polyps (average 23.6 mm height, n=4 polyps) that are more robust (average 5 mm, n=4 polyps) than seen in P. heliodiscus, but with almost no development of the coenenchyme, unlike as in P. mutuki or other closely related species. As well, this specimen is from 3 meters depth, a shallower depth than usually seen for P. heliodiscus.
RMNH Coel 40512 ( Figure 6D) is a small 'intermediae' colony consisting of four polyps that are squat and robust (average width 8.3 mm, n=3 polyps, height approximately same as width) with large oral discs (average 12 mm in diameter, n=3 polyps) with no tentacles visible and a large oral opening.
Previous records. NA.
Remarks. The morphology of these specimens do not clearly match any described species from the central Indo-Pacific. In particular, specimen RMNH Coel 40512 is different than any other zoantharian previously observed by the first author. However, it is unknown if fixation has resulted in degradation of fine scale structures (e.g. tentacles, which are absent), but the specimen is clearly a zoantharian due to sand encrustation in body wall.  Description. This zooxanthellate species was described in detail recently by Ryland and Lancaster (2003). Superficially similar in appearance to P. mutuki, externally the species can be distinguished by its short tentacles (length <20% of oral disk) and subtidal distribution, compared to primarily intertidal P. mutuki, which also has longer tentacles (~45% of oral disk) (Ryland and Lancaster 2003).
Sizes of specimens agree well with specimens seen in other localities (average polyp heights 11.3 mm and 17.0 mm for each specimen, range 7-20 mm; average width 3.9 mm and 4.4 mm for each specimen, range 3.5-5.5 mm; n=2 specimens of 8 and 5 polyps, respectively). Depth of collected specimens (12 and 35 m) also fits well with the description of this species as primarily subtidal in the original description, and from data in Okinawa, Japan (e.g. Reimer 2010).
Previous records. Palythoa heliodiscus has been reported from Australia (Ryland and Lancaster 2003) and is likely widespread in the Indo-Pacific (Ryland and Lancaster 2003 and references within), as well as Japan (Reimer et al. 2006b), Palau (Reimer et al. 2014a), while P. toxica Walsh & Bowers, 1971 has been reported from Hawai'i.
Remarks. We have identified all specimens here as "cf." as in situ images ( Figure 8) there are two different morphotypes. One morphotype matches with Palythoa heliodiscus, with a brown oral disk with no patterns (Figures 8A, B), while the other morphotype's polyps have either green or purple oral disks with various semi-irregular patterns, as well as blue/gray or light orange tentacles ( Figures 8C, D). Based on data from Okinawa and Australia, both of these morphotypes are almost identical asides from the oral disk coloration and small but consistent differences in ITS-rDNA (T. Nishimura and J.D. Reimer, unpubl. data) that may be either intraspecific or interspecific. Thus, it is still uncertain if the green/purple morphotype is an undescribed species or not (Reimer et al. 2014a).
Furthermore, the overall morphology of the green/purple morphotype closely resembles P. toxica from Hawai'i, and whether these Indonesian specimens are P. toxica or P. heliodiscus, and if these two species are synonyms needs to be ascertained before any formal description occurs. In situ images and further DNA sequences are therefore needed from future specimens. Photographic records. NA. Description. This specimen superficially resembles zooxanthellate Palythoa sp. yoron sensu Shiroma and Reimer (2010) with its very well developed coenenchyme and 'intermediae-immersae' morphology ( Figure 9B). However, there are some differences between this specimen and P. sp. yoron from Okinawa. The current specimen consists of two large portions of colonies consisting of >50 polyps, while P. sp. yoron usually is found in very small colonies of <10 polyps. As well, P. sp. yoron consists of a very well developed coenenchyme from which all individual polyps partially emerge, while the current specimen appears to consist more of large robust polyps that have merged together at many locations, but not at others, giving the specimen the appearance of P. tuberculosa from the top, and often of P. mutuki from side angles. On the other hand, P. sp. yoron has an appearance, although intermediate between P. tuberculosa and P. mutuki, unique to and of itself. Polyps' height (when not merged) is approximately 7.0 mm, and average width is 7.3 mm (n=10 polyps). Thus, for now, this specimen is identified as P. aff. tuberculosa. For details on P. tuberculosa, refer to the relevant species section below.
Remarks. This specimen is unlike any other previous specimen observed in the field or museums by the first author. Unfortunately, as it was collected in 1930, attempts to acquire utilizable DNA sequences able to distinguish this specimen's affinity were unsuccessful, and identification was made on gross morphology alone.
Specimens in this study averaged 4.7 mm in polyp diameter (n=29 specimens), ranging from 2 to 8 mm. One specimen, RMNH Coel 40553, was notable for its very small polyps (average diameter 2.4 mm, n=10 polyps). Other colonies ranged from 3.1 to 6.5 mm in average diameter, similar to previous reported sizes. All specimens were 'immersae'. Generally, morphology fit well within the accepted range of P. tuberculosa (see Table 1 in Hibino et al. 2013), although some specimens' polyps were somewhat smaller than previously observed. These smaller sizes may also be partly due to preservation methods.

Remarks.
It is highly likely this species is the senior synonym of P. caesia Dana, 1846(Hibino et al. 2013, described from Fiji and commonly reported from Australia (Burnett et al. 1997). This species is also part of the P. tuberculosa-P. mutuki species complex (Reimer et al. 2007b, M. Mizuyama and J.D. Reimer unpubl. data October 11, 2003;station BER.14, lighthouse northeast Pulau Panjang Island, East Kalimantan, Berau Islands (02°23'14"N, 118°12'34"E), October 9, 2003;station BER.24, southeast Pulau Samama, East Kalimantan, Berau Islands (02°07'51"N, 118°20'23"E), October 15, 2003. Description. The type species of the azooxanthellate genus Sphenopus, this species has an Indo-West Pacific distribution (Reimer et al. 2012b). Uniquely for the order, species in this genus are unitary (not colonial), and usually free-living, as they are not attached to substrate, and instead embedded in sand or loose gravel/substrate ( Figures  10A, B). Individuals can often grow to large sizes (for zoantharians); up to several cm in both length and polyp diameter. Taxonomic examination of this genus is quite limited, with only two recent studies (Soong et al. 1999, Reimer et al. 2012b, both of which clearly state that further research is needed to more clearly understand this group. Specimen RMNH Coel 40506 consists of seven polyps, with an average height of 24.4 mm (range 18.5 to 30 mm), and an average width of 8.4 mm (range 6 to 11 mm). The non-peduncle portions of the polyps are 15-20 mm in height, with the remainder made up of peduncle.
Specimen RMNH Coel 40506 has some polyps (five of seven) somewhat different in morphology from RMNH Coel 40509 and other Naturalis S. marsupialis specimens from the Indian Ocean. These polyps have regularly spaced small round "tubercles" (approx. 1 mm in diameter) on the upper half of their scapus arranged in vertical lines (n=8-14 vertical tubercle lines on each polyp, with 6-10 tubercles per line), making this portion of the polyp appear furrowed. As well, polyps have a small, stubby "peduncle" (2 to 5 mm in width) that is not attached to any hard substrate, intermediate between S. marsupialis with its completely rounded bottom end and S. pedunculatus with its long, attached peduncle. For now, we identify these specimens as S. marsupialis as their peduncles were not attached to the substrate, but it is clear more examination of these specimens is needed.
Specimen RMNH Coel 40509 consists of two polyps of different sizes, with the smaller one being 16 by 5 mm, and the larger one 24 by 15 mm. Both polyps have no peduncle and are tapered. Both polyps are somewhat rugged on their outer surface, with no discernable tubercles, and have intermittent (=not one clear stripe) small darker vertical patterns in between the capitulary ridges only on the top 3-5 mm of the oral end of polyps.
Previous records. This species has been reported from many locations in the Indo-West Pacific, including Taiwan (Soong et al. 1999) and Brunei Darussalam (Reimer et al. 2012b).
Remarks. Specimen RMNH Coel 40506 may be similar to a putative undescribed Sphenopus species mentioned in Soong et al. (1999) from Taiwan based on its smaller size. Description. This azooxanthellate species was originally described from the Philippines, and has not been reported in the literature for over 80 years, excepting two brief mentions in Reimer et al. (2012b). Easily discernable from other Sphenopus species by the presence of a 'foot' (=peduncle) that is attached to substrate (e.g. small rocks).

Sphenopus pedunculatus
The two specimens here varied in length from 33 to 62 mm in polyp length, and had a width between 9 to 11 mm (polyp head). The "swollen", non-peduncle part of the polyp was between 15 to 20 mm in height, with the remainder of the length made up of the peduncle, which was between 0.5 to 3 mm in width. RMNH Coel 40507 polyps were generally smooth in appearance, while the upper portions of polyps of RMNH Coel 40510 were somewhat rugged, with small round tubercules 0.5 mm in diameter roughly arranged in vertical lines. The spaces between these small tubercules were colored a much darker color than the remainder of the polyps' outer surfaces. The peduncle of specimens and images ( Figure 10C) are much thinner and longer than the sketch in Hertwig (1888). However, so few data are available for this (and other Sphenopus species) that currently nothing is known about intraspecific variation, and for now, we group these two specimens within this species.
Previous records. This species was originally described from the Philippines, but has not been mentioned in recent literature (except for Reimer et al. 2012b), and hence very little is known on its distribution or ecology.
Remarks. It is unknown as to whether the peduncle is a morphological characteristic that forms only when there is a hard substrate available, and this needs to be investigated to confirm this is truly a different species from S. marsupialis.  Figures 12A, B  Description. As originally and previously described (Di Camillo et al. 2010), this azooxanthellate, colonial species is found as an epibiont on hydrozoans, particularly Plumularia habereri Stechow, 1909. In this study, this species consists of only one morphotype, with a gray to brown scapus, and reddish-brown oral disk and tentacles ( Figure 12B). The appearance matches well with the morphotype of the species observed by Di Camillo et al. (2010).
In this study, measurements are only available for two specimens, with polyps averaging 2.4 mm in height and 2.1 mm in width. These data also fit well with Di Camillo et al. (2010), who mention polyp heights of 2-5 mm, widths of 1.6 to 3 mm, with 32 tentacles and mesenteries.
Remarks. This morphotype differs from the other known morphotype of the species (sensu Carlgren 1934) associated with this binomen, which is yellow in coloration. The original description of the species from Sagami Bay, Japan by Lwowsky (1913) was of a "gray, sandy" morphotype, but this was preserved in formalin, and thus could be either morphotype discussed here, or even a different one altogether. Phylogenetic analyses have shown subtle differences of sequences of specimens within this species (Sinniger et al. 2008), indicating that taxonomic revision may be needed in the future for this species group.
Description. Similar to H. gracilis above, this azooxanthellate, colonial species is found as an epibiont on Plumularia habereri. As described in Di Camillo et al. (2010; as Parazoanthus sp.), this species has much smaller polyps than H. gracilis, forming colonies only on the main branch(es) of Pl. habereri colonies. Polyps are much less in-crusted than H. gracilis. The Pl. habereri colonies hosting this species are much bigger than those with H. gracilis, as shown by (Di Camillo et al. 2010). Red scapus with yellow tentacles, 22 to 24 tentacles slightly longer than oral disk diameter ( Figure 12C).
Remarks. This undescribed species was informally and well described by Di Camillo et al. (2010) as "Parazoanthus sp.". Specimens and DNA sequences are needed to properly describe this species. Figures 12D, 13 Specimens examined. NA.
Description. Azooxanthellate. Polyps well free and clear of coenenchyme. Outer surface of polyps covered with dense incrustation of irregularly sized sand particles, reminiscent of Microzoanthus sp. Oral disk semi-translucent with dark, almost black coloration, except for oral opening, which is much lighter in color. 40 to 50 tentacles, at least as long as oral disk diameter, with same blackish coloration as oral disk, with terminal 1/4 whitish in coloration. Colonies attached to non-living substrate. Specimen RMNH Coel 40469 is apparently a fragment of a whole colony, while the photographic record shows a colony of approximately 50 polyps arising from a common coenenchyme ( Figure 14A). The single specimen had polyps averaging 6.8 mm in length (n=3) and 3.1 mm in width (n=6).  (Figure 15). Spermonde Archipelago (9), Moluccas (14).

Distribution. Regions recorded in this study
Previous records. None, although similar undescribed specimens have been photographed in the Philippines (P. Poppe, pers. comm.), and collected from Okinawa, Japan (Reimer, unpubl. data), indicating a potential West Pacific distribution.
Remarks. This species is similar in appearance but different in coloration to T. onoi from the Galapagos and west coast of Central and South America. Figures 14C, D, 15 Specimens examined. NA.
Description. With only a single photographic record available, even an informal description of this undescribed species is limited. Asides from yellow coloration, this Figure 15. Distribution of Terrazoanthus species from specimens and photographic records from this study. Terrazoanthus sp. 1 specimens in red, and Terrazoanthus sp. 2 in green. Region numbers correspond to locations given in species' information. Boxes indicate presence of specimens (with or without photographic records), while circles indicate only photographic records. Overlapping symbols indicate the same region. species is outwardly similar to Terrazoanthus sp. 1 above. Polyps appear to be more crowded than in Terrazoanthus sp. 1, with 40 to 54 yellow tentacles longer than oral disk diameter ( Figure 14D).
Remarks. This species has been traded in the aquarium trade as "yellow polyps", and is thought to be distributed primarily in Indonesia, yet no museum specimens exist, preventing this species from being formally described. Colonies often appear to be intermixed with Zoanthus spp. colonies in shallow water (J.D. Reimer, pers. obs.). Although undescribed, this putative species has been placed with the genus Terrazoanthus based on DNA sequences acquired from aquarium trade polyps (Sinniger et al. 2005, Reimer andFujii 2010 November 11, 1996;RMNH Coel 40768, Snellius Expedition, Pulau Bo Islands, Halmahera Sea, collected on October 5, 1930. Photographic records (n=1). Station BER.30, north of Lighthouse 1 Reef, south of Pulau Derawan, East Kalimantan, Berau Islands (02°16'02"N, 118°14'23"E), October 22, 2003. Description. Azooxanthellate, epibiotic on Keroeides sp., polyps approximately the same height as width (approximately 1-3 mm), connected by coenenchyme visible on the outer surface of the octocoral colony. Polyps numerous, placed between smaller octocoral polyps, pale yellow in coloration, with outer surface of polyps slightly reddish in color similar to host octocoral. Tentacles relatively short, approximately half of the oral disk diameter, also pale yellow, and approximately 20 in number ( Figure 16A).
Specimen RMNH Coel 40766 is larger than RMNH Coel 40768 (polyp average width 2.6 mm vs 1.6 mm, respectively). However, the latter specimen is quite old (from the original Snellius Expedition) and this difference may be due to fixation methods.
Previous records. NA.

Remarks.
Only two specimens and one photographic record of this undescribed species exist. However, these records are each from different expeditions, and it is reasonable to expect that this species is at least distributed in the Banda and Celebes Seas.
Recently, many different genera in the family Parazoanthidae have been described based on a combination of epizoitic relationships and phylogenetic analyses (e.g. Sinniger et al. 2010Sinniger et al. , 2013. However, no parazoanthids have been reported in association with Keroeides, and therefore currently it is impossible to place these specimens and records into a genus without both further examination of specimens combined with DNA sequence data.  width 2.1 mm, n=8 polyps) and do not protrude much from the coenenchyme, with polyp height approximately same as width. Polyps and coenenchyme are heavily encrusted, and golden yellow-brown in color. Coenenchyme forms a thin sheath over the antipatharian surface. Capitulary ridges not clearly discernable. Polyps form semi-regular vertical rows over short distances of the antipatharian (e.g. approx. 5 cm), but with no observable pattern for the entire colony ( Figure  16C). Colony encrusts the top approximately 1/2 of the Cirripathes specimen; starting approximately 15 cm from the bottom holdfast. The Cirripathes colony's proximal tip appears to be broken off and missing.
Remarks. This species may belong to genus Antipathozoanthus, which was described recently by Sinniger et al. (2010) and includes species from both the Atlantic and Indo-Pacific, with reports of specimens also from the Red Sea (Reimer et al. 2014b). It is likely several undescribed Antipathozoanthus species are present in the Indo-Pacific, as only one Antipathozoanthus species from the Galapagos has been formally described. In situ images and DNA sequences are needed to formally describe this species.  Figures 18A, 19 Specimens examined. NA.
Past records. Previously, similar specimens have been reported from Japan (Sinniger et al. 2008) and the Red Sea (Reimer et al. 2014b).
Past records. NA.
Remarks. The only sponge-associated Parazoanthus species formally described from the Indo-Pacific are P. elongatus McMurrich, 1904 from the west coast of South America and New Zealand (Sinniger and Haussermann 2009) and P. darwini Reimer & Fujii, 2010 from the Galapagos. Thus, no similar species have been reported from past or recent zoantharian work in surrounding CIP regions, and therefore it is likely that these specimens constitute an undescribed species. Although Parazoanthus has recently been taxonomically redescribed (Sinniger et al. 2010), and the species now only encompasses sponge-associated species, the genus is still paraphyletic and taxonomic revision is needed. DNA sequences from undescribed species are needed to correctly place specimens such as these into the correct clade. Description. This putative azooxanthellate species is similar in size to Parazoanthus sp. 2 above, with polyps of average 6.1 mm in height (range 2.5 to 10 mm; n=2 colonies) and average width of 3.2 mm (range 2 to 4 mm). Some small dark incrustations visible on lower half (=aboral) of polyps' scapus. Approximately 20 capitulary ridges, indicating tentacle counts of approximately 40. Polyps range from cream (RMNH Coel 40525) to tan (RMNH Coel 50545) in color when preserved. Polyps arise from a well-developed stoloniferous coenenchyme in rows, with most found along the upper and outer edges of flat, paddle-shaped sponges ( Figures 18C, D). No polyps found on the lower 'foot' or 'stalk' of sponges.
Past records. NA. Remarks. Similar in size to Parazoanthus sp. 2 above, we have included these two specimens as a separate putative species in this study. This is partly due to specimens being in association with a different sponge species (compare Figures 18B, C), as host specificity may differ between species (e.g. Crocker and Reiswig 1981;Swain and Wulff 2007 Description. Originally described from the Philippines. Azooxanthellate. Polyps of this species often grow at the outer bends of the zig-zag shaped tubes, and combined with polyps' smaller size and thin coenenchyme ( Figure 20A), colonies appear to be much less crowded than as seen in E. aff. illoricatus ( Figure 20B).
Polyps of specimens in the RMNH collection are generally less than 1 mm in diameter, and never more than 2 mm, and of approximately equal height. Coenenchyme generally light gray in color, oral disk and tentacles semi-translucent brown. Tentacles in images 20-22 in number, much thinner than as seen in Epizoanthus aff. illoricatus below, with orange or white colored proximal tips, longer in length than oral disk diameter. The two deeper specimens (RMNH Coel 40533 and 40546) have highly developed thin coenenchymes covering the entire worm tubes' surface, and are both dark black in color. On the other hand, the shallower specimens had some unitary polyps, and colonial polyps were often in clusters of two or three with poorly developed coenenchyme.
The morphological characters and dimensions observed in the specimens in this study agree well with the original description by Tischbierek (1930).
Past records. Originally described from Manila, and subsequently reported from Taiwan (Reimer et al. 2013a), New Caledonia (Sinniger 2006), Palau (Reimer et al. 2014a), and Osprey Reef, Australia in the Coral Sea (Lindsay et al. 2012), indicating a western Indo-Pacific distribution. Found from specimens in this study as shallow as 20 m, and as deep as 90 m, similar to depths reported at Osprey Reef (=82.5 m) (Lindsay et al. 2012).
Remarks. Until this report, any Epizoanthus spp. on a zig-zag shaped eunicid worm was recorded as E. illoricatus. However, from the preliminary analyses here, it appears that there may be two or more species within this group. Thus, previous records must be treated with caution. 125°11'E), October 23, 1994; west Pulau Siladen, Bunaken, North Sulawesi (01°38'N, 124°46'E), May 2, 1998.
Description. Azooxanthellate. As E. illoricatus above, obligate epibiont on eunicid worms. Polyps of this putative species are at least twice as big in diameter as E. illoricatus (average 2.1 mm, compared with a maximum of 2 mm for E. illoricatus), and many times bigger in terms of volume. Additionally, both specimens have brown coenenchyme and scapus, different from the light gray coenenchyme and brownish oral disk reported for E. illoricatus ( Figure 20B). In situ images show colonies with cream, brown, red-brown, orange-brown or tan coenenchyme and scapus, often with white tentacles that are comparatively shorter and thicker than in E. illoricatus. The coenenchyme of this putative species is much more developed than E. illoricatus, with polyps arising from not only bends of the zig-zag shaped eunicid tube, but also from other locations. The result is a colony that has a higher density of polyps than E. illoricatus. In E. illoricatus, often the zig-zag shape of the eunicid tube is visible between polyps, but this is rarely the case in E. aff. illoricatus ( Figure 20B). Figure 21. Distribution of Epizoanthus species from specimens and photographic records from this study. Epizoanthus illoricatus specimens in red, and Epizoanthus aff. illoricatus in green. Region numbers correspond to locations given in species' information. Boxes indicate presence of specimens (with or without photographic records), while circles indicate only photographic records. (Figure 21): East Bali (5), northeast Sumba (6), Spermonde Archipelago (9), Moluccas (14), Lembeh Strait (17), Bunaken (18).

Distribution. Regions recorded in this study
Past records. NA. Remarks. Although the two specimens here were found at deeper depths (75 to 190 m), numerous photographic records show that this species and E. illoricatus have an overlapping depth range. Examination of DNA sequences combined with detailed morphological analyses should help clear up the relationship between this putative species and E. illoricatus, although preliminary molecular analyses show differences between the two groups (H. Kise and J.D. Reimer, unpubl. data). It is likely records and museum specimens identified as E. illoricatus from the central Indo-Pacific include both types mentioned in this study.

Discussion
Examination of the Naturalis zoantharian collection resulted in 24 species being identified, 12 from suborder Brachycnemina and 12 from Macrocnemina. While by no means an extensive collection, with most specimens from Indonesia, these results indicate that the Central Indo-Pacific waters are at least as diverse in numbers of species, genera, and families as surrounding regions of Australia, Singapore, and Japan. In Australia, an examination of the brachycnemic shallow water zoantharians of the Great Barrier Reef indicated the presence of eight species (Burnett et al. 1997), while in Okinawa, 12 brachycnemic species have been previously reported (Reimer 2010), and in Taiwan 10 species (Reimer et al. 2013a). These previous reports did not include macrocnemic species, but from the brachycnemic results alone, Indonesia does appear to have zoantharian species diversity at least as high as Okinawa, one of the most well examined regions. Finally, as many macrocnemic species live in deeper areas or in caves and other less-examined ecosystems (Sinniger et al. 2013), we expect the number of zoantharian species in the shallow waters of Indonesia to be higher than the initial estimate in this study, and further investigations should confirm this idea.
The discussion of total numbers of shallow water zoantharian species is clearly still in the preliminary stages given the lack of focused sampling throughout most regions of the world, as well as the continuous discovery of new species and genera (Reimer et al. 2007a, Sinniger et al. 2010, Fujii and Reimer 2011. Still, the initial species numbers from this study should provide a basis upon which future zoantharian studies can build on. Furthermore, it should not be forgotten that the previous reports listed above from other Indo-Pacific regions were all conducted by zoantharian specialists collecting specimens in the field, while the Indonesian specimens in the Naturalis collection constitute only part of a broad sampling effort of benthic invertebrates without the participation of any zoantharian specialists. Thus, our prediction that the total number of shallow water zoantharian species in Indonesia will be considerably higher than reported in this study is almost certainly accurate, particularly given the recent discovery of widespread yet cryptic zoantharian species from coral reef environments Reimer 2011, 2013) not yet reported from Indonesia.
Further supporting the possibility of Indonesia harboring a diverse zoantharian fauna is the fact that the specimens examined in the Naturalis collection are primarily from eastern Indonesia (e.g. Sulawesi and Banda Sea, Fig. 1), with few or no specimens from other regions such as the islands of Java and New Guinea, and only one locality in the Philippines and Papua New Guinea. Future collection efforts must be focused on these unexamined regions if we are to obtain a clear understanding of zoantharian diversity in the CIP. Additionally, the deep sea (>200 m depth) has been recently speculated to harbor much undiscovered zoantharian diversity (Sinniger et al. 2013) and yet in this study only three of the zoantharian species were found in waters >50 m in depth. Exploring the deeper waters in the Indonesian region will undoubtedly result in further discoveries.
Of the 24 total species listed in this study, at least nine (and perhaps up to 12 if Palythoa spp. are included) are likely undescribed species. Some, such as Terrazoanthus sp. 2, have been known for years in the global aquarium trade, yet still no museum specimens exist, and thus we cannot formally describe them within this manuscript. Without formal descriptions and a clear understanding of species, future conservation work cannot proceed effectively, and immediate taxonomic efforts should focus on the obtaining of specimens and a formal description of this species. Similarly, many photographic records exist for Neozoanthus sp., yet no specimens are in the Naturalis collection.
Three other species that are almost certainly undescribed species are Parazoanthus sp. 1, Parazoanthus sp. 2 and Parazoanthus sp. 3. Until now, only two sponge-associated Parazoanthus species has been formally described from the Pacific, and none from sub-tropical or tropical waters. Five other species, Hydrozoanthus sp. 1, Hydrozoanthus sp. 2, Parazoanthidae sp. 1, Parazoanthidae sp. 2, and Terrazoanthus sp. 1, are also very likely to be undescribed species, but with only photographic records, or one or two specimens existing for these species, additional specimens and molecular data are needed to properly describe them.

Conclusions
In conclusion, this study provides a starting point for zoantharian research in the Coral Triangle. We were able to discern 24 different morphological species based on specimen examination combined with photographic records. However, based on recent previous research, phylogenetic analyses of specimens would likely result in somewhat different results due to both high levels of intraspecific morphological variation in some species (Burnett et al. 1997, Reimer et al. 2004) and also morphological convergence between other phylogenetically distinct species (Sinniger and Haussermann 2009). These previous studies demonstrate how difficult it often is to properly identify zoantharian species without molecular data.
Furthermore, this study demonstrates that the central Indo-Pacific likely harbors very high levels of zoantharian diversity, as the numbers of putative species from this study (24) include a large number of undescribed species, and total numbers are as high or higher than previously reported for any other region.
Finally, it is hoped that this study can serve as a template for the study of other understudied coral reef benthos in the Coral Triangle. In this study, past photographic records proved to be invaluable in aiding species identification, and understanding species distributions. Thus, while museum collections should remain the key tool in taxonomic and biogeographic research (Rainbow 2009, Rocha et al. 2014, archived in situ images can provide additional information.