Nautiloids are a charismatic group of marine molluscs best known for their rich fossil record, but today they are restricted to a handful of species in the family Nautilidae from around the Coral Triangle. Recent genetic work has shown a disconnect between traditional species, originally defined on shell characters, but now with new findings from genetic structure of various Nautilus populations. Here, three new species of Nautilus from the Coral Sea and South Pacific region are formally named using observations of shell and soft anatomical data augmented by genetic information: N. samoaensis sp. nov. (from American Samoa), N. vitiensis sp. nov. (from Fiji), and N. vanuatuensis sp. nov. (from Vanuatu). The formal naming of these three species is timely considering the new and recently published information on genetic structure, geographic occurrence, and new morphological characters, including color patterns of shell and soft part morphology of hood, and will aid in managing these possibly endangered animals. As recently proposed from genetic analyses, there is a strong geographic component affecting taxonomy, with the new species coming from larger island groups that are separated by at least 200 km of deep water (greater than 800 m) from other Nautilus populations and potential habitats. Nautilid shells implode at depths greater than 800 m and depth therefore acts as a biogeographical barrier separating these species. This isolation, coupled with the unique, endemic species in each locale, are important considerations for the conservation management of the extant Nautilus species and populations.

Conservation, deep-sea, Nautilidae, Nautilus, taxonomy

During the mid-19^th century period of active study of nautilus taxonomy, little was known about their natural history or even the number of species of extant Nautilus Linnaeus, 1758. From the 1980s onward, diverse research concerning the biology and taxonomy of living nautiluses, including aspects of reproduction, inter- and intraspecific variation in shell morphology, distribution, and genetic variation, has shed light on this charismatic clade (Saunders 1981, 1987; Saunders and Landman 1987; Ward 1987; Woodruff et al. 1987; Wray et al. 1995; Bonacum et al. 2011; Kröger et al. 2011; Sinclair et al. 2011; Vandepas et al. 2016; Ward et al. 2016; Combosch et al. 2017; Zhang et al. 2021). This research has profoundly changed Nautilus taxonomy and significantly reduced the number of recognized species, often based on single specimens found in Australian waters and otherwise without provenance (Saunders 1987; Ward 1987). It further revealed that there are far more shell characters that can be utilized for taxonomy than previously realized. For example, Ward (1987) cited differences in shell coloration and shell size at maturity, proposing that Nautilus populations from Vanuatu and Fiji were sufficiently dissimilar to the type of N. pompilius Linnaeus, 1758 to warrant taxonomic differentiation. However, no new species or subspecies were formally defined in that publication.

The number of valid species accepted in the genus Nautilus has remained unsettled, even as new genetic information has added to our understanding of the variability within and between Nautilus species. Currently, the species that have been described are differentiated based upon their geographic habitat, including the three species described here (Fig. 1).

Figure 1. 

Geographic range map of Nautilus and Allonautilus.

The three “universally” accepted modern nautilid species of 21^st century taxonomists are Nautilus pompilius (the type species), N. macromphalus G.B. Sowerby II, 1849, and N. stenomphalus G.B. Sowerby II, 1849. A fourth, N. belauensis Saunders, 1981, has been disputed, and is not currently supported by phylogenetic studies; it does not show significant morphological differences from all other extant species (Vandepas et al. 2016; Combosch et al. 2017; Tajika et al. 2021). It has been placed as a junior synonym of the type N. pompilius by Vandepas et al. (2016) but accepted as a valid species elsewhere (e.g., Jereb 2005). Similarly, controversial has been the acceptance of Nautilus repertus Iredale, 1944, which genetic studies have linked with N. belauensis (Vandepas et al. 2016; Combosch et al. 2017; Saunders et al. 2017; Tajika et al. 2021), and which is currently considered a taxon inquirendum (WoRMS 2021). Additionally, the type designation for N. pompilius was recently designated to come from a neotype in Ambon, Indonesia (Nikolaeva et al. 2015), which most significantly brings into question the validity of N. pompilius in the Philippines, where a subspecies has already been described by Habe and Okutani (1988) as N. pompilius suluensis. However, recent genome-wide genetic studies did not distinguish this (a single) Indonesian sample from the Philippine Nautilus population (Combosch et.al 2017). Future work on the relationship of these two range populations of Nautilus would surely improve our understanding of their genetic relationships.

Another significant taxonomic change in Nautilus systematics was the definition of the new genus Allonautilus (Ward and Saunders 1997). From the time of its first description, Nautilus scrobiculatus Lightfoot, 1786 was recognized to be morphologically different from all other nautiluses in shell shape, shell ornament, and shell coloration. However, with the first collection of living specimens of this taxon (Saunders and Davis 1985), significant differences in shell ultrastructure (the presence of a thick periostracum anatomically dissimilar to the homologous layer in all other extant nautilids), and anatomical and color differences of the fleshy hood were demonstrated. Later, dissection of preserved specimens revealed further differences from known species of Nautilus in the respiratory and reproductive systems (Ward and Saunders 1997). In consequence, Nautilus scrobiculatus, was removed from Nautilus and designated instead as the type species of a new genus, Allonautilus Ward & Saunders, 1997. Genetic studies (Wray et al. 1995; Sinclair et al. 2007; Bonacum et al. 2011; Sinclair et al. 2011; Vandepas et al. 2016; Combosch et al. 2017) have since confirmed significant genetic differences between A. scrobiculatus and the various other species placed in Nautilus, further supporting the designation (Bonacum et al. 2011; Vandepas et al. 2016; Combosch et al. 2017; Tajika et al. 2021).

Early phylogenetic studies based predominantly or exclusively on mitochondrial DNA data indicated major problems with conchological-defined Nautilus species and instead identified three geographically distinct clades (Bonacum et al. 2011; Vandepas et al. 2016) and concluded that most Nautilus species may not be valid. The subsequent application of genome-wide RAD-Seq data (Combosch et al. 2017) greatly increased the phylogenetic resolution among and within these three major clades and confirmed the presence of genetically distinct subclades among South Pacific Islands in New Caledonia (N. macromphalus), Vanuatu (N. vanuatuensis sp. nov.), Fiji (N. vitiensis sp. nov.), and American Samoa (N. samoaensis sp. nov.), of which the three latter species are described further in this paper. Interestingly, some phylogenetic analyses by Combosch et al. (2017) indicated significant genetic differences between Fiji and American Samoa, while a population genetic analysis (STRUCTURE) did not (Fig. 2), which may have been due to the limited sample size from both locations. Ongoing follow-up analyses using expanded datasets and recently generated genome assemblies (Zhang et al. 2021; Huang et al. 2022) will further clarify species boundaries and provide additional population genetic insights into the drivers of differentiation among island populations.

Figure 2. 

Phylogenetic relationships of nautilids based on maximum-likelihood analysis of 18,595 concatenated SNPs with IQtree as in Combosch et al. (2017). Numbers on nodes indicate ultrafast bootstrap resampling percentages (>50) based on 1000 replicates–asterisks indicate full support (* = 100%). Bootstraps values were replaced on intraclade nodes as follows: “-” for bootstraps values >50 – <90, “+” for bootstrap values ≥90. Colors correspond to geographic populations as indicated in the map on the bottom left. Species other than Nautilus pompilius are indicated by name at the tip of the tree. The STRUCTURE plots show the posterior probability for individual assignments of samples to different genetic clusters in clade-specific analyses. Distinct colors indicate different genetic clusters as inferred by STRUCTURE (not populations as on the tree and map).

In parallel with the various genetic studies, ecological observations of Nautilus populations have been undertaken (Saunders 1984; Dunstan et al. 2011a; Barord et al. 2014; Ward et al. 2016), including understanding of maximum habitat depths imposed by shell implosion of juveniles and adults. Coupled with the lack of planktonic dispersal of newly hatched nautiluses (Ward 1987), this indicated the potential for geographic isolation of populations surrounded by water depth deeper than implosion depths of 800 m and, thus, the potential for allopatric speciation. The behavior of adult nautiluses observed through ultrasonic tracking (Carlson et al. 1984; Ward et al. 1984; Dunstan et al. 2011b; Ward et al. 2016) is consistent with the hypothesis that nautiluses have limited dispersal ability.

The work of Combosch et al. (2017) provided genome-wide evidence consistent with the view that geographic distances and depths readily crossed by invertebrates and vertebrates with planktonic larval stages are effective barriers to migration for nautiluses. Our new mapping of Nautilus and Allonautilus populations (Fig. 1) with observable morphological differences as well as information from genetic studies suggest that 200 km of deep water (800 m or greater) between separated populations is sufficient to effectively prevent gene flow. These results documenting the genomic structure in Nautilus populations living on the isolated archipelagos of American Samoa, Fiji, and Vanuatu, as well as new morphological information presented below, led to the conclusion that the nautiluses at each of these archipelagos are part of distinct populations recognized as new species. While three of these have heretofore been placed in N. pompilius (e.g., Ward et al. 1977; Saunders 1987; Ward 1987; Woodruff et al. 1987; Bonacum et al. 2011), in this paper we define each as a new species of Nautilus, based on shell characters and hood morphology as well as previously published genome-wide population genetics results.

We use a combination of previously published data, as well as measurements from both museum collections and from newly collected specimens to produce the complete character assemblage data presented here. The combination of morphological characters included: shell coiling descriptors, shell ornament, umbilicus, and hood morphology. One aspect of extant shell morphology that has not been used at the species level is shell decoration, including pattern, pigment hue, and percent of shell covered by pigment. Ward (1987) proposed that the latter (as viewed from the side) was more variable between species than within species and was the basis of the proposal that various species at the time accepted as N. pompilius in American Samoa, Fiji, and Vanuatu could be discriminated from the types by their different percentages of pigment cover. Here we use the percent of the shell covered with pigment, as viewed from the side, modified from the methods introduced by Ward (1987). When combined with other characters, pigment cover becomes useful for species level distinction. Specifically, we combine pigment cover with size at maturity. Shell color pattern as a discriminative character was separately presented by Ward (1987), who listed data taken from photographs of captured specimens, and then measured with a manual planimeter. We use essentially the same method here but use the far more discriminating software, ImageJ. In both cases, however, the photographs with the shell striping percentage of a side view of the entire shell was manually obtained using the NIH ImageJ program to separately measure the area of the entire shell, and then measuring that area from a side view that was pigmented, by manually tracing each stripe with the “Irregular circular area” tool. The patterns and the actual area of shell they were compared to (which excluded the area under the black region beneath the hood, and the umbilical region), is shown in Fig. 3.

Figure 3. 

Example of method used to determine percentage of pigmentation on lateral portion of shell in N. samoaensis sp. nov.

Additionally, a new character used here for species-level definition and useful for interspecies discrimination is mean shell diameter of mature specimens, as in Ward (1987) and Saunders (1987). All measures presented here come from photographs of either empty shells that are part of museum collections or living animals that were photographed and soon after released. In some cases, photographs and videos were also obtained from Nautilus populations in American Samoa, Fiji, and Vanuatu using baited remote underwater video systems (BRUVS) as described by Dunstan et al. (2011a) and Barord et al. (2014). Shell sizes were taken from various sources where size measurements were presented; these sources are listed in the captions of the various diagrams portraying the measured data.

The populations of nautiluses within American Samoa, Fiji, and Vanuatu have already been effectively isolated from each other because of warm, surface seawater temperatures, depth implosion limits below 800 m, and their nektobenthic lifestyle, one of endless foraging just above the bottom, and thus within a few meters of the benthic environment at most. Here, we combined morphological characteristics with previously published population genomic results to newly describe each population of Nautilus within these archipelagos as a unique species. The characters that showed the most differentiation between the species included multiple color pattern traits (i.e., percent of shell pigmented in matures specimens and striping patterns) and, while not yet quantified, the actual “hue” of pigment in freshly caught specimens.

Fig. 7 shows the shell decoration and hood ornamentation of each of the species described here. Shell size at maturity also has discerning potential among some of the species’ groups (Fig. 8). Nautilus samoaensis sp. nov., N. vitiensis sp. nov., and N. vanuatuensis sp. nov. represent the easternmost range of extant nautilids and provide further validation for the decision to regulate all nautiluses (family Nautilidae) within appendix II of the Convention on International Trade in Endangered Species (CITES) given their morphological and genetic differences.

Figure 8. 

Shell diameter and shell pigmentation relationship of N. vitiensis sp. nov., N. samoaensis sp. nov., and N. vanuatuensis sp. nov. compared against each other as well as other Nautilus species and populations.

For comparative purposes, if we include the New Caledonian populations of nautiluses, N. macromphalus, with the three new species described here we can draw further inferences on the relationship of distinct species of nautiluses. The strongest morphologic similarity among the four species we designate as endemic to this region is between N. macromphalus and what the population assigned to N. pompilius from the Vanuatu archipelago, N. vanuatuensis sp. nov. They are clearly identified by the umbilical region, although indistinguishable in all other studied characters (Ward 1987). These two species are morphologically similar, as well as being geographically proximate to one another and, as noted above, also share the behavioral trait of commonly swimming at night into depths shallower than inhabited by any other Nautilus species. However, some of these traits must be adaptational, as these two species are not each other’s closest relatives (Fig. 2).

The third of the four species living in what we call an “South Pacific” biogeographic region (sensu Combosch et al. 2017), Nautilus samoaensis sp. nov. resembles both N. macromphalus and N. vanuatuensis sp. nov. in size and the relative amount of coloration on the shell but differ markedly in shell color “pattern.” It is sister to N. vitiensis sp. nov. and while these two species were ambiguously distinguished in prior genetic work, their reciprocal monophyly and the morphological differences outlined above, lead us to designate them as distinct species here.

Individuals of N. vitiensis sp. nov. in Fiji have never been seen at night, and the exceptionally low amount of pigmentation in this species may be due to an overall, deeper-water habitat, as discussed above. In any event, N. vitiensis sp. nov. has among the lowest degree of shell coloration of all currently defined species. New analyses of previously published shell formation temperatures of oxygen isotopes from shell and septal samples (Tajika et al. 2022) are consistent with the growth of juvenile species at depths below 300 m; these data, however, tell us nothing about habitation depths of the non-growing mature specimens.

Compared to N. macromphalus and N. vanuatuensis sp. nov., N. samoaensis sp. nov. has a unique species-level character: it has the most striking and readily identifiable ornamental pattern of any known Nautilus or Allonautilus species. Specimens captured with baited traps and observed through BRUVS observations made in 2013 at depths of 250–350 m all show coloration patterns on the middle portion of the phragmocone, which is the rare (for Nautilus), single character trait that identifies the species: a complex change in stripe direction (compared to striping found in any other Nautilus). Different as it is from the Vanuatu–New Caledonian species couplet, N. samoaensis sp. nov. still remains closer in observable shell characteristics to those two South Pacific species than it does to the geographically closer nautiluses of Fiji. The latter are smaller, and their shells bear a significantly different ornamentation than those of the other three species identified within this biogeographic region.

These data may be skewed, as all sampled populations of nautiluses report that males vastly outnumber females in traps, which is related to size at maturity. Nevertheless, the size at maturity remains a powerful character, and correlates well with genetic data. In fact, as we show here, the currently accepted and newly proposed species herein, with the sole exception of N. pompilius, can be statistically separated using a combination of the characters of mature shell size, percentage of shell covered by pigment, the morphology of the shell umbilical region, the morphology of the shell surface at a growth line scale, the thickness of the periostracum (which is dependent on shell surface morphology), and the morphology and color of the hood. Undoubtedly, other characters will be discovered when detailed dissection of the accepted species is finally attempted, including comparing the morphology of the radula and jaws. Nevertheless, even with the characters at hand, we show here that populations known to be separated by at least 200 km of deep water maintain significant genetic differences as well as measurable morphological differences, which is consistent with our conclusion that the populations of American Samoa, Fiji, and Vanuatu merit elevation to separate species. The use of morphological characters, shell pattern, and, to a lesser extent, shell size, have additional power when regulating and enforcing current trade regulations for all nautiluses. The fact that we were able to combine the morphology and genetics to differentiate these species provides a foundation for managers and other officials to begin to efficiently identify distinct species of nautilus shells that may come through as trade products.

The three species, N. vitiensis, N. samoaensis, and N. vanuatuensis represent populations of nautiluses on the easternmost edge of the overall habitat range of Nautilus. The designation of these three populations as distinct species provides insight into evolutionary radiation of the genus and clarification for future conservation practices.

We thank Michael Vecchione for assistance with curated specimens within the Smithsonian Institution. Associate Editor Jiří Frank, Amane Tajika, and an anonymous reviewer provided comments that helped refine this work. This manuscript was published by a grant from the Wetmore Colles Fund of the MCZ.