Alviniconcha spp. are iconic hydrothermal vent snails, yet almost no life history or population structure information exists for species within this genus. Here, we present biological trait data for the original hairy snail, Alviniconcha hessleri. We investigated reproductive periodicity, larval developmental mode, and allocation to reproduction in adults, including population size-frequency, oocyte size, and protoconch-I size. This species occurs in multiple vent fields from the Mariana back-arc, which differ in geothermal conditions, resulting in vent effluents with differing chemical compositions. Alviniconcha hessleri hosts diverse chemosynthetic bacteria (Stein et al. 1988; Breusing et al. 2020, 2022), and it is possible that the combination of symbiont identity and chemical and thermal conditions could impact the productivity of the symbiosis and affect reproductive partitioning in the form of oocyte size.
While not the focus of this study, the unexpected discovery of morphologically hermaphroditic individuals within this species is the second such occurrence within the genus. Castel et al. (2024) reported a high frequency of intersex individuals among Alviniconcha kojimai that were initially visually identified as female. Histological sections of A. kojimai individuals revealed mostly female gametic tissue, with the sporadic appearance of male tissues at various stages of development, which led Castel et al. (2024) to suggest that these individuals may represent an intermediate step in evolution between gonochorism and hermaphroditism. Interestingly, the histological sections of intersex individuals of A. hessleri in the present study more frequently revealed small patches of female tissue appearing sporadically throughout male tissue. Given the apparent lack of gametic organization and previous work revealing no significant relationship between snail size and sex within A. hessleri (seemingly ruling out sequential hermaphroditism; Warren 2019), it may be that the sexual systems of these species are indeed experiencing evolutionary change. The discovery of intersex individuals within A. hessleri and A. kojimai in particular is intriguing as these species are suggested to be the most closely related within the genus (Johnson et al., 2015) despite inhabiting geographically isolated back-arc systems.
Oocyte development indicates continuous reproduction and planktotrophy
Examination of histological sections revealed asynchronous oogenesis across individuals at all sites. Ovaries contained all three stages of oocyte development within a single female and all oocyte size-frequency distributions tended to be unimodal. Our results are consistent with continuous gametogenesis within females of A. hessleri, as in the majority of hydrothermal vent molluscs studied (Fretter 1989; Gage and Tyler 1991; Gustafson and Lutz 1994b; Tyler and Young 1999; Kelly and Metaxas 2007; Tyler et al. 2008; Matabos and Thiebaut 2010; Poitrimol et al. 2024). While the closest relative to Alviniconcha, Ifremeria nautilei, displays discontinuous gametogenesis and the novel use of a brooding pouch (Reynolds, 2009), other vent gastropod taxa, such as Lepetodrilus species, show continuous oogenesis with female gonads containing all oocyte developmental stages (Kelly and Metaxas 2007; Tyler et al. 2008; Poitrimol et al. 2024). Continuous or semi-continuous gametogenesis have also been noted in other dominant vent taxa, such as the bivalve Calyptogena magnifica(Berg 1985) and the alvinellids Alvinella pompejana (Faure et al. 2007), Paralvinella pandorae and P. palmiformis (McHugh 1989). In the hydrothermal environment where there is a mostly continuous energy supply, continuous reproduction is considered a strategy to take advantage of active venting. Maximizing reproductive output and larval dispersal would ensure the continual establishment of new populations at other vents, protecting against the unpredictability of geothermal activity, and thus food availability.
Maximum oocyte diameter was 93 µm in A. hessleri, which is on the lower end of reported oocyte sizes for vent gastropods. At least four Lepetodrilus species have maximum oocyte sizes between 80–100 µm (Tyler et al. 2008), while several other gastropod taxa have a maximum size of 100–150 µm (Kelly and Metaxas 2007; Tyler et al. 2008; Matabos and Thiebaut 2010; Marticorena et al. 2020; Poitrimol et al. 2024). Oocyte size is often used as an indicator of life history, where larger eggs provisioned with more yolk imply direct development or lecithotrophy, and smaller oocytes indicate planktotrophy and the potential for wider dispersal (Gustafson and Lutz 1994b). However, given that these relationships between life history characters have mostly been observed in shallow species, it is unknown whether these patterns hold in the deep-sea.
While extensive studies linking oocyte size and mode of development do not exist for vent species, a comprehensive study of the larval biology of the caenogastropod genus Conus revealed a significant correlation between egg size and pelagic duration (Kohn and Perron 1994). Out of 64 Conus species, oocyte size for planktotrophic species ranged from 125–425 µm, one instance of lecithotrophy was found at an oocyte size of 390 µm, and the oocyte size of direct developers ranged from 470-1,000 µm (Kohn and Perron 1994). Of the hydrothermal vent gastropod taxa studied, most appear phylogenetically constrained to exhibit lecithotrophic development (Berg, 1985; Gustafson and Lutz, 1994; Lutz et al., 2008). Despite this, a feeding planktotrophic larva is believed to be the ancestral condition for deep-sea gastropods (Bouchet and Warén 1994) and oocyte size within this study suggests planktotrophy for A. hessleri. However, oocyte size should not be taken in isolation to indicate larval life history, but other parameters such as protoconch size, as well as the comparison of similar characters in congeners should be considered for a more definitive interpretation.
Recognized sources of error involved with histological processing may have influenced some our measuments, and should be considered in future studies which rely on our data for comparative purposes. For instance, we may have underestimated maximum oocyte size and relative abundance of vitellogenic oocytes, as larger oocytes were less likely to meet our requirements for inclusion. Due to their larger size, an obvious oocyte boundary was not always evident, and occasionally yolk droplets could be observed in isolation, suggesting that some oocytes may have lost their integrity during processing. Histological processing can also shrink tissues up to ~ 20% (Tyler et al. 2008). These sources of error combined highlight that histologically-based oocyte sizes may be an underestimate of actual size, further increasing the importance of using other characters to inform interpretations of larval life history among gastropod species.
Protoconch-I size estimate indicates excessive larval shell degradation
Few options exist for characterizing larval life history when larvae are not accessible and cannot be cultured, and most studies use protoconch size to estimate larval trophic strategy and planktonic duration (Bouchet and Warén 1994; Kohn and Perron 1994). Unfortunately, degradation of the larval shell in vent gastropods can occur during exposure to the acidic hydrothermal fluids, which can increase the difficulty of estimating protoconch size. This is particularly true for Alviniconcha species, where even the smallest specimen post-metamorphosis shows damage at the apex. As the juveniles grow, a calcareous plug is secreted to fill the apex and protect the snail. It is unknown whether the plug approximates the dimensions of the original protoconch, therefore we only used the smallest, most intact specimens available to measure protoconch size. In A. hessleri, the protoconch-teleoconch transition is not obviously demarcated by a sculptural change in the shell. Thus, the initial whorl of the shell can be used as an approximation of protoconch-I, as it would have needed to encompass the visceral mass of the developing embryo, including the majority of the endogenous yolk. Here, we presented a hypothetical measure of protoconch-I to investigate whether degradation of the larval shell occurs in a predictable way and to help inform on the life history of A. hessleri.
The hypothetical protoconch-I measurement for A. hessleri was ~ 675 µm, which is larger than in the original description by Okutani and Ohta (1988), who reported a protoconch of ~ 500 µm. In addition to this inconsistency, a study examining zooplankton community composition in the North Pacific Subtropical Gyre genetically identified the presence of A. hessleri larvae within 200 and 500 µm size classes from zooplankton tows located north of Hawaii (Sommer et al. 2017). Thus, hatching size (protoconch-I size) of A. hessleri larvae are most likely smaller than the sizes reported in this study and in the original description by Okutani and Ohta (1988). This indicates that more than one whorl degrades post-metamorphosis, and estimates of protoconch-I size in adults are not a reasonable substitute for intact larval shells. Beyond this, the capture of A. hessleri larvae roughly 6,000 km away from their only known habitat (Sommer et al. 2017) also supports the conclusion that this species has a long larval period, typical of planktotrophic larvae, and rasies important questions about the transport and vertical position of larvae within the water column.
Given that many vent species exhibit lecithotrophic development, vent larvae have been assumed to remain at depth during development (Gustafson and Lutz 1994a), which may facilitate retention within vent systems and explain the limited range of some species. Previously modelled dispersal patterns have indicated that A. hessleri larvae in the Mariana back-arc likely disperse using cold bottom currents, with larvae dispersing at a depth of 3000 m predicted to be 20% more likely to arrive at another vent destination within the back-arc than those dispersing at the surface (Breusing et al. 2021a). However, the record of A. hessleri larvae occurring thousands of kilometers away from their only known habitat (Sommer et al. 2017) suggests that the larvae of this species may utilize stronger mid-oceanic or upper-layer oceanic currents. Adams et al. (2012) speculated that vent species of the East Pacific Rise could potentially use upper ocean countercurrents to form a loop facilitating transport and return of larvae in the Pacific Ocean, however examples of this are lacking. Evidence of larvae dispersing in mid-water, and sometimes even in the photic zone, exists for a few comparatively shallow vent and cold seep species, such as alvinocaridid shrimp (Herring 2006), limpet snails (Yahagi et al. 2017) and bathymodiolus bivalves (Arellano et al. 2014). But, for a species occurring only at depths greater than 3000 m such as A. hessleri, this degree of vertical transport (and even greater horizontal transport) resulting in continued successful recruitment is unprecedented. While evidence from this study and published literature imply planktotrophy for this species, increased sampling effort is needed to clarify the larval ecology of this species.
Population size structure indicates irregular recruitment events
The population size structure distribution of A. hessleri varied across sites. All sites except Voodoo Crater differed from a normal distribution, with a strong bimodal population at Sequoia. Distributions other than unimodal may suggest periodic recruitment, which is difficult to reconcile with the continuous gametogenesis observed in all individuals in this study, but may be explained by periodic recruitment. For instance, the size structures observed herein are most likely due to environmental influences on larval recruitment, including directional currents and variability in venting activity. Breusing et al. (2021) used genetic data to assess larval dispersal patterns, which revealed that both our northern sites (Illium and Burke) contain source populations for the central and southern sites (Sequoia, Voodoo Crater, and Perseverance). The same work also indicates that Sequoia, Voodoo Crater and Perseverance receive larvae from southern vent fields outside this study, Forecast, Snail and Pika. This may explain why we see younger populations at Sequoia in particular, as it may be receiving larvae from both directions, leading to stronger recruitment pulses. Similarly, the inferred dispersal directionality could explain why there are fewer recruitment events at Illium, which also appeared to contain the oldest population, with snail sizes significantly larger than at most other sites.
Oocyte size may be conserved beyond a reduced chemical threshold
Given that sites along the Mariana back-arc differ substantially in parameters such as temperature, reduced hydrogen sulfide and methane concentrations, differences in fluid chemistry may influence the symbiosis between A. hessleri and its bacterial endosymbionts, thus impacting the available energy for host allocation to reproduction. However, the relationship between temperature and reduced chemical content was not as straightforward as expected, as the hotter sites of Burke and Illium did not consistently have higher reduced chemical concentrations. Illium had the lowest hydrogen sulfide and methane of all sites, suggesting that this vent may be older and is starting to senesce. Interestingly, the only site to show significant differences in A. hessleri characteristics was Illium, displaying significantly larger shell height and significantly smaller oocyte sizes, suggesting that the unique fluid chemistry at Illium may be influencing the biology of A. hessleri.
With a temperature of ~ 29°C, Illium should be optimal for the A. hessleri holobiont, as warmer fluid chemistries are typically associated with higher quantities of reduced chemicals (Van Dover 2000). However, Illium had a remarkably low hydrogen sulfide level of 13 µM, whereas all other sites exceeded 150 µM. Because the dominant symbiont type in A. hessleri is thiotrophic (Stein et al. 1988; Breusing et al. 2022), one interpretation of our results is that the symbiosis is indeed less productive at this site due to a lack of hydrogen sulphide, and that this may have downstream effects for provisioning to oocytes by the host. However, differential allocation of resources to reproduction could also take the form of variation in total fecundity, which should be accounted for in future studies for species within this genus. A confounding variable is that the Illium population was also significantly larger in body size based on shell height than all other sites (except Voodoo Crater), so it is possible that this population is older and may be decreasing provisioning to eggs due to age. However, gametogenic maturity did not differ among sites, and individuals at Illium still had a proportion of vitellogenic oocytes greater than 50%, except for one individual.
Research examining spatial variation in oocyte size is mixed for symbiotic vent gastropods.Kelly and Metaxas (2007) found that symbiotic Lepetodrilus fucensis from senescent vents on the Juan de Fuca Ridge showed a decrease in oocyte size, quantity, and proportion of vitellogenic oocytes compared to those from more actively venting sites. In contrast, three species of Lepetodrilus from the East Pacific Rise(Tyler et al. 2008) and two from the west Pacific(Poitrimol et al. 2024) showed no spatial variation in oocyte size. Thus, it might be that as long as a certain reduced chemical threshold is met, allocation to reproduction in vent gastropods is highly conserved. Two other factors could also convey resilience in reproductive allocations in A. hessleri across differing chemical environments: 1. its ability to graze as well as receive nutrition from symbionts, 2. its diverse endosymbiont population (the most diverse in the genus; Breusing et al., 2022). Given the relatively plastic nature of some life history traits despite claims of phylogenetic constraints in the Gastropoda, further reproductive studies on Alviniconcha and sister taxa are required to confirm whether different vent chemistries can impact host reproductive effort.