Extreme longevity in a deep-sea vestimentiferan tubeworm and its implications for the evolution of life history strategies


The deep sea is home to many species that have longer life spans than their shallow-water counterparts. This trend is primarily related to the decline in metabolic rates with temperature as depth increases. However, at bathyal depths, the cold-seep vestimentiferan tubeworm species Lamellibrachia luymesi and Seepiophila jonesi reach extremely old ages beyond what is predicted by the simple scaling of life span with body size and temperature. Here, we use individual-based models based on in situ growth rates to show that another species of cold-seep tubeworm found in the Gulf of Mexico, Escarpia laminata, also has an extraordinarily long life span, regularly achieving ages of 100–200 years with some individuals older than 300 years. The distribution of results from individual simulations as well as whole population simulations involving mortality and recruitment rates support these age estimates. The low 0.67% mortality rate measurements from collected populations of E. laminata are similar to mortality rates in L. luymesi and S. jonesi and play a role in evolution of the long life span of cold-seep tubeworms. These results support longevity theory, which states that in the absence of extrinsic mortality threats, natural selection will select for individuals that senesce slower and reproduce continually into their old age.


Within the depths of the ocean, there are extraordinarily long-lived animals including the octopus Graneledone boreopacificia, found up to 2000 m deep, with the longest brooding period known for any animal (Robison et al. 2014) and black corals (Antipatharia) more than 4000 years old (Roark et al. 2009). Rockfish, a group with closely related species distributed across a broad depth gradient, show a trend of increasing life span from 12 to 205 years with increasing maximum depth of occurrence from 37 to 874 m depth (Cailliet et al. 2001). A similar trend of slower growth rates with depth is apparent in the octocorals (Andrews et al. 2002). The slower pace of life at depth can be explained best by the effects of temperature on metabolism as ambient temperature decreases steadily in deep waters (McClain et al. 2012).

One of the longest-lived animals on the planet, the cold-seep vestimentiferan tubeworm Lamellibrachia luymesi (Polychaetea: Siboglinidae), exhibits lower mortality rates and a longer life span than predicted by metabolic scaling according to body size and temperature alone (McClain et al. 2012). Even in a study that created universal scaling laws specifically for deep-sea organisms across the globe using the metabolic theory of ecology, the data from L. luymesi were recognized as outliers (McClain et al. 2012). Another Gulf of Mexico cold-seep tubeworm, Seepiophila jonesi, would also have appeared as an outlier in the study had its life span been included in the dataset (Cordes et al. 2007). The question remains how widespread these life history traits are among the cold-seep vestimentiferans and if longevity continues to increase with depth in this group.

Evolutionary theories offer an alternative approach to understanding life span beyond unifying equations that scale with body mass and environmental factors. The premise of longevity theory is that senescence is allowed to evolve because natural selection has only a weak influence on genes that contribute to deterioration with aging. Selection has only a weak effect on these genes because most members of natural populations are killed before they can reach old age and experience the detrimental effects of these genes (Williams 1957; Kirkwood and Austad 2000). If extrinsic mortality threats are reduced, however, selection on these genes is stronger and the organisms that possess genes that allow them to live longer and reproduce viable offspring later in life become the most successful (Austad 1993). The deep sea offers a more stable environment and generally has fewer large predators than other biomes, decreasing the frequency of mortality events.

Cold seeps provide an additional level of stability beyond that provided by the majority of the deep sea through the steady release of fluids from the seafloor for thousands of years at a single seep (Roberts and Aharon 1994; Cordes et al. 2009; Ingram et al. 2010). These fluids are rich in methane and/or hydrogen sulfide and bring a reliable energy source to the microbes inhabiting the seeps, whether they be free-living or in symbiosis (Paull et al. 1985; Cordes et al. 2009). Cold-seep tubeworms are entirely reliant on internal bacterial symbionts for their nutrition and live an effectively autotrophic lifestyle (Fisher 1990). In addition to the geological stability of the cold seep’s sulfide supply, tubeworms biologically enhance this source of sulfide by facilitating microbially mediated sulfur recycling (Cordes et al. 2005). Microbes in the sediment couple anaerobic methane oxidation to sulfate reduction using sulfate released from the tubeworms’ posterior “roots,” and this process produces hydrogen sulfide to be taken up by the tubeworm and used by its symbionts (Cordes et al. 2003; Dattagupta et al. 2006). When mediated by L. luymesi, this sulfide-replenishing process has been modeled to supply the tubeworm’s metabolic needs for at least 250 years (Cordes et al. 2005). Given this stable energy supply, it is possible for L. luymesi to reach life spans over 250 years and S. jonesi over 300 years (Bergquist et al. 2000; Cordes et al. 2007). Escarpia laminata is another chemosynthetic bacteria-hosting vestimentiferan tubeworm of the Gulf of Mexico that inhabits seeps between approximately 1000 and 3300 m depth, but little is known about this species and its life history. This study aims to estimate the life span of E. laminata and determine if this deeper-dwelling tubeworm species exhibits longevity similar to its shallower relatives.

Material and methods

Collection and in situ growth measurement

To investigate age and life history, the growth of E. laminata tubeworms was measured in situ using methods developed for L. luymesi (Bergquist et al. 2000). Six aggregations of E. laminata were stained with Acid Blue #158 (Fig. 1) and seven different unstained aggregations were sampled with the Bushmaster Jr. collection device (Bergquist et al. 2003) in June 2006. In June 2007, the six stained aggregations were located again and collected. The total length of each tubeworm was measured from the anterior end of the tube to the posterior portion that tapered to 2 mm in diameter. The white anterior portion of the tube (new growth above the blue dye line) was measured as the growth between staining and collection, approximately 1-year growth. In total, six populations were collected from Atwater Valley 340 (27.64° N, 88.36° W, 2180 m), two from Alaminos Canyon 818 (26.18° N, 94.62° W, 2745 m), one from Alaminos Canyon 601 (26.39° N, 94.51° W, 2425 m), and four from Green Canyon 852 (27.11° N, 91.19° W, 1410 m). Growth was measured from 356 total stained tubeworms and standardized to an annual growth rate based on staining and collection dates, and all 13 Bushmaster-collected aggregations including an additional 1046 unstained tubeworms were used as reference populations for the population simulations. In addition to the stained tubeworms, five E. laminata individuals banded with zip ties and ID tags in 1992 at Alaminos Canyon 645 were found alive in 2007 and re-imaged to calculate their average annual growth rate over 15 years by dividing all new growth by 15 years.

Fig. 1

The tubes visible in the left half of the picture are unstained, natural E. laminata, and the tubes on the right show the conspicuous Acid Blue #158 stain that was applied to distinguish new growth

Growth model and individual simulations

The measured growth rates from the stained E. laminata individuals were fit to the negative exponential equation g = ae -bL derived from L. luymesi growth rates (Fig. 2) in the program JMP (SAS Institute Inc 2013; Bergquist et al. 2000; Cordes et al. 2007). In this model, g is annual growth (cm year−1), L is length (cm), and a and b are constants that were calculated during the regression (a = 2.401 and b = 0.0367). To create an error term for the model, another nonlinear regression was performed on the positive residuals from the growth equation. The residuals were best fit by another negative exponential equation ε = ce -dL, where c and d are constants (c = 2.102 and d = −0.0294). The probability of whether or not an individual would grow in a given year was also found to be correlated to the size of the tubeworm, similar to the cold-seep tubeworm species Seepiophila jonesi (Cordes et al. 2007). After calculating proportions of E. laminata showing nonzero growth in bins of 20 individuals (ordered by size), the probability of annual growth was best fit by the equation p g  = x + ye -zL where x, y, and z are constants (x = 0.264, y = 3.751, and z = −0.0880). If a uniformly distributed random number between zero and one was greater than the probability p g , the individual tubeworm in simulation would increase its length by the size-dependent amount g + N(0,ε) centimeter that year.

Fig. 2

a One year’s growth for all collected E. laminata plotted against the size of the tubeworm measured. The solid line is the best fit negative exponential equation g = ae -bL (R 2 = 0.1302), and the dashed lines show the 95% confidence interval around the equation. b Comparable measured growth data from the species L. luymesi and the negative exponential equation that best fit that dataset, reproduced from (Cordes et al. 2007)

To estimate the mean age of the tubeworms in each population, the number of years required to reach the mean length of that population’s tubeworms’ lengths was estimated by recording what year the growth simulation exceeded this mean length. The same method was applied to estimate the age of the largest individuals collected from each population. To account for the fact that these E. laminata individuals may be the largest not because they are significantly older but because they grow faster than average, a second growth simulation for these sizes was run using an increased growth rate. These maximum growth simulations used higher percentile values for a and b in the size-dependent annual growth rate equation (greater than the 50th percentile values used in the average simulation and based on the size of the population, percentiles given in Table 1), and tubeworms grew every year without exception. In each growth scenario (average growth to mean length, average growth to max length, max growth to max length), the number of years required to exceed the recorded length was averaged across 10,000 runs.

Table 1 Individual simulation results

Population simulations

The average individual growth model was extended with mortality and recruitment rates to construct a population-wide simulation in order to estimate the ages of tubeworms in these collections while capturing more of the variability in their growth rates.

The rate of removal of individuals from the population was estimated by counting the number of empty tubes that were collected within the populations. Using in situ chitin degradation rates of Riftia pachyptila (a hydrothermal vent species of tubeworm) tube material and adjusting for the temperature difference between cold seeps and hydrothermal vents (Ravaux et al. 2003), it was estimated that the tubes can stand for 3 years in a seep environment after the death of the worm and still be solid enough (up to 50% degraded by microbial activity) to be collected and measured, allowing for the conversion of this count into an annual rate. The probability of a mortality event given a tubeworm’s size was modeled by calculating the proportion of empty tubes collected across 5 cm size bins. The mortality rate for E. laminata was found to be extraordinarily low at 0.67% of the population dying annually, and this rate did not change with tubeworm length.

In order to successfully recruit to a seep, vestimentiferans require exposed hard substrate and hydrogen sulfide. Substrate surface area is rapidly occupied by the first recruits, and the low mortality rate of cold-seep vestimentiferans reduces the incidence of space being made available for a late recruit by the death of an individual already settled at that site. Additionally, hydrogen sulfide levels in the water column are depleted over time, and thus recruitment ceases after a few decades (Bergquist et al. 2002; Cordes et al. 2003). This is evident in the complete absence of small tubeworms (<10 cm) in any of the populations sampled for this study and challenges the application of recruitment models developed for other species. The density-dependent model R = Gf published by Kohyama and Takada (1998) was adapted here for E. laminata (Kohyama and Takada 1998). Instead of calculating recruitment rate from the smallest size classes, the maximum recruitment pulse was estimated from the most common size class. To calculate recruitment, the number of individuals in the most common 1 cm size class for each collected E. laminata population was correlated with the areal coverage of the aggregation in square meter. The largest possible sampling area was 0.283 m2 (the size limit of the sampling device), and the smallest sampling area measured was 0.006 m2 due to the small (n = 5) population of tubeworms. There was a significant linear relationship (P < 0.0001) with a slope of 76.36 ind m2. The size class was assumed to represent 1-year cohort (R = 76.36 ind m2 year−1), and sensitivity tests showed no significant effect on the estimated age of the simulated aggregation or the quality of the fit of the simulation to the collected data from varying this parameter between 7.64 and 152.7 ind m2 year−1.

Each simulated tubeworm in the population grew at the rates defined by the average growth model. Individuals in the simulated population would be removed from the population if a uniformly distributed random number between zero and one was less than the mortality rate 0.0067 ind year−1. The population simulation continued to recruit the number of new individuals generated by a normally distributed random number with mean R until the population reached carrying capacity (defined as the total number of individuals collected in the reference population). Each year of the population simulation, the size distribution of the simulated tubeworm population was compared to the size distribution of the collected population using a log-likelihood test. The simulation year with the lowest log-likelihood test statistic was recorded as the estimated age of the population, and the age was then averaged across 10,000 simulation runs for each population.


The E. laminata specimens measured displayed a similar trend to L. luymesi in decreasing annual growth with increasing size, although E. laminata tubeworms of comparable size grow less than half as fast (Fig. 2). According to the average individual growth simulation, an E. laminata individual with an anterior length of 50 cm is predicted to be 116.1 (sd = 19.4) years of age (Fig. 3), older than L. luymesi and S. jonesi individuals of the same size that have been estimated to be 21 and 96 years old, respectively (Cordes et al. 2007). At the fastest growth rates, corresponding to the 99th percentile of measured rates, a 50-cm long E. laminata would be 19.8 (sd = 1.7) years of age. At greater lengths, the age disparity becomes larger with E. laminata estimated to reach 1 m anterior length after 1232 (sd = 85.6) years, 67.9 (sd = 2.3) years at maximum growth rates, compared to 68 years for L. luymesi and 605 years for S. jonesi to reach the same size. Although L. luymesi tubeworms frequently reach lengths longer than 1 m, 50 cm is a more ecologically relevant comparison for the smaller species E. laminata and S. jonesi. In the average growth simulations, the age at which tubeworms could attain the mean length of each population ranged from 15 to 266 years from the smallest to the largest mean length (Table 1). The age estimates for the largest individuals from each population ranged from 77 to 15,834 years using the average growth model and from 17 to 1059 years when the maximum growth model was applied (Table 1).

Fig. 3

Growth was simulated for 1500 years for 10,000 individual tubeworms and the maximum, minimum, and averages sizes for each year were recorded. After 57 years, the average size of the simulated individuals was equal to the mean length of all collected E. laminata (42 cm), marked with vertical and horizontal dotted lines. It took 812 years for even the fastest growing tubeworms in the simulation to reach the 99th percentile length of all tubeworms collected (100 cm), shown with the larger vertical and horizontal dotted lines

The range of growth rates measured from the banded E. lamintata individuals was 0.67 to 2.67 mm year−1 with an average of 1.38 mm year−1 over the 15-year period. Although the total length of the zip-tied individuals was not measured, these growth rates fall within the values predicted by the model for moderate to large tubeworms and suggest that the stained tubeworms’ single year of growth was a representative sample of rates exhibited by E. laminata over their lifetime. At the average growth rate of 1.38 mm year−1 from the 15-year banding data, it would take an individual 362 years to grow to 50 cm and 725 years to reach 1 m in length, both of which within the range of estimates from the other methods.

The averages for each collected population range from 21 years to over 9000 years (Table 2). In 10 of the 13 simulated populations, the mean length of tubeworms averaged across all simulation runs in the best-fit year was within 10 cm of the measured mean length for that collected population. Population-level simulations have the benefit of capturing more observed growth variation, but in four simulated populations the average log-likelihood comparisons between the modeled populations and the actual size-frequency histograms were significantly different (P < 0.05) in even the best-fit years. The nine simulated populations that were not significantly different from collected populations are marked with asterisks in Table 2. The goodness-of-fit test between real and simulated data was most likely to indicate similar distributions when the reference population had a normal distribution and small range between the largest and smallest individual. The individual- and population-level approaches, despite their limitations, both indicate that the larger E. laminata individuals reach ages in excess of 250 years.

Table 2 Population simulation results


These findings all support the conclusion that E. laminata tubeworms are much older than similarly sized L. luymesi tubeworms. The explanation for this disparity is not immediately clear given the close phylogenetic relationship of these species and their ecological similarity. A full population simulation is more likely to capture the true growth rates of all the collected tubeworms whereas individual aging methods may be inaccurate if the largest tubeworms experienced anomalously fast growth rates. Calculating the age of large tubeworms is also hampered by the asymptotic nature of the growth model, which introduces a high degree of error in the estimates of the age of the largest E. laminata individuals. The two oldest population simulation ages are more than 25 times larger than the next oldest estimate, suggesting that these results are outliers and not a reliable evaluation of E. laminata’s life span. Even removing these outliers, E. laminata individuals are reliably predicted to live in excess of 250 years, and may occasionally attain far greater ages if given a steady supply of hydrogen sulfide and oxygen in the absence of predation.

L. luymesi is found on the upper Louisiana slope between 300 and 950 m deep; whereas, E. laminata occurs only on the lower slope below 1000 m (Cowart et al. 2014), and this difference in their depth ranges may potentially explain their life span disparity. Some studies have found decreasing rates of metabolism with increasing depth of occurrence in some deep-sea species (Childress et al. 1990; Drazen and Seibel 2007), and a lower resting metabolic rate can lead to a longer life span (Atanasov 2005). Although the metabolic rate of E. laminata has not been measured directly, there was a significant relationship between depth and growth rate where tubeworms were more likely to grow less than the model predicted the deeper they were collected (between 1409 and 2746 m). This relationship only held for larger individuals, as the relationship was only significant when the high degree of variability from small individuals (<30 cm) was removed from the dataset. Observed trends of decreased metabolism with depth are most likely due to the direct effect of temperature on metabolism, rather than depth per se (Childress et al. 1990). All E. laminata collection sites were between 4.2 and 4.3 °C, and the average temperature difference between these and L. luymesi sites is less than 4 °C. According to the metabolic theory of ecology, E. laminata’s life span would be only 1.34 times longer than L. luymesi’s if the difference were due to temperature alone (McClain et al. 2012).

The interactions between E. laminata’s metabolism, life span, and mortality can also be investigated through published allometric relationships. When calculating the predicted mortality using terms averaged across fish and invertebrates (McCoy and Gillooly 2008), E. laminata has a mortality rate 15 times lower than the equation predicts given the tubeworms’ average dry mass and the ambient temperature of the seeps at which it was collected (avg. 4.2 °C). Although hydrothermal vent and cold-seep tubeworms may be subject to non-lethal plume-cropping (Bergquist et al. 2007; Cordes et al. 2010), lethal predation of tubeworms has not been demonstrated (Bergquist et al. 2003). E. laminata does have a blood-sucking parasite in the polychaete Protomystides sp., and this small polychaete has also been found inside the empty tubes of dead tubeworms, but whether these small parasites can cause mortality at appreciable rates is unknown (Becker et al. 2013). Thus L. luymesi, S. jonesi, and E. laminata all have similarly low mortality rates. This low rate of individual turnover places E. laminata in the category with L. luymesi and S. jonesi as anomalously long-lived deep-sea species according to the metabolic theory of ecology (McClain et al. 2012). These calculations give increased importance to extrinsic mortality rate in increasing animals’ life spans.

At more than 250 years old, E. laminata achieves a life span that exceeds other longevity records such as 177-year-old Galapagos giant tortoises (the longest-lived land vertebrate) and 211-year-old bowhead whales (the longest-lived mammal) (Deweerdt 2012). The marine clam Arctica islandica remains the oldest non-colonial animal known with an inferred age of 507 years (Ridgway and Richardson 2011), but given the uncertainty associated with estimating the ages of the longest individuals, there may be large E. laminata tubeworms alive in nature that live even longer.


  1. Andrews A, Cordes E, Mahoney M (2002) Age, growth and radiometric age validation of a deep-sea, habitat-forming gorgonian (Primnoa resedaeformis) from the Gulf of Alaska. Hydrobiologia 101–110

  2. Atanasov AT (2005) The linear alometric relationship between total metabolic energy per life span and body mass of poikilothermic animals. Biosystems 82:137–142. doi:10.1016/j.biosystems.2005.06.006

    Article  PubMed  Google Scholar 

  3. Austad SN (1993) Retarded senescence in an insular population of Virginia opossums (Didelphis virginiana). J Zool 229:695–708. doi:10.1111/j.1469-7998.1993.tb02665.x

    Article  Google Scholar 

  4. Becker EL, Cordes EE, Macko SA et al (2013) Using stable isotope compositions of animal tissues to infer trophic interactions in Gulf of Mexico lower slope seep communities. PLoS One 8:e74459. doi:10.1371/journal.pone.0074459

    Article  PubMed  PubMed Central  Google Scholar 

  5. Bergquist D, Williams F, Fisher C (2000) Longevity record for deep-sea invertebrate. Nature 403:499–500

    CAS  Article  PubMed  Google Scholar 

  6. Bergquist D, Urcuyo I, Fisher C (2002) Establishment and persistence of seep vestimentiferan aggregations on the upper Louisiana slope of the Gulf of Mexico. Mar Ecol Prog Ser 241:89–98. doi:10.3354/meps241089

    Article  Google Scholar 

  7. Bergquist DC, Ward T, Cordes EE et al (2003) Community structure of vestimentiferan-generated habitat islands from Gulf of Mexico cold seeps. J Exp Mar Bio Ecol 289:197–222. doi:10.1016/S0022-0981(03)00046-7

    Article  Google Scholar 

  8. Bergquist D, Eckner J, Urcuyo I et al (2007) Using stable isotopes and quantitative community characteristics to determine a local hydrothermal vent food web. Mar Ecol Prog Ser 330:49–65. doi:10.3354/meps330049

    Article  Google Scholar 

  9. Cailliet G, Andrews A, Burton E et al (2001) Age determination and validation studies of marine fishes: do deep-dwellers live longer? Exp Gerontol 36:739–764. doi:10.1016/S0531-5565(00)00239-4

    CAS  Article  PubMed  Google Scholar 

  10. Childress JJ, Cowles DL, Favuzzi JA, Mickel TJ (1990) Metabolic rates of benthic deep-sea decapod crustaceans decline with increasing depth primarily due to the decline in temperature. Deep Sea Res Part A Oceanogr Res Pap 37:929–949. doi:10.1016/0198-0149(90)90104-4

    CAS  Article  Google Scholar 

  11. Cordes EE, Bergquist DC, Shea K, Fisher CR (2003) Hydrogen sulphide demand of long-lived vestimentiferan tube worm aggregations modifies the chemical environment at deep-sea hydrocarbon seeps. Ecol Lett 6:212–219. doi:10.1046/j.1461-0248.2003.00415.x

    Article  Google Scholar 

  12. Cordes EE, Arthur MA, Shea K et al (2005) Modeling the mutualistic interactions between tubeworms and microbial consortia. PLoS Biol 3:e77. doi:10.1371/journal.pbio.0030077

    Article  PubMed  PubMed Central  Google Scholar 

  13. Cordes EE, Bergquist DC, Redding ML, Fisher CR (2007) Patterns of growth in cold-seep vestimenferans including Seepiophila jonesi: a second species of long-lived tubeworm. Mar Ecol 28:160–168. doi:10.1111/j.1439-0485.2006.00112.x

    Article  Google Scholar 

  14. Cordes EE, Bergquist DC, Fisher CR (2009) Macro-Ecology of Gulf of Mexico cold seeps. Annu Rev Mar Sci 1:143–168. doi:10.1146/annurev.marine.010908.163912

    Article  Google Scholar 

  15. Cordes EE, Becker EL, Fisher CR (2010) Temporal shift in nutrient input to cold-seep food webs revealed by stable-isotope signatures of associated communities. Limnol Oceanogr 55:2537–2548. doi:10.4319/lo.2010.55.6.2537

    Article  Google Scholar 

  16. Cowart DA, Halanych KM, Schaeffer SW, Fisher CR (2014) Depth-dependent gene flow in Gulf of Mexico cold seep Lamellibrachia tubeworms ( Annelida, Siboglinidae ). Hydrobiologia 736:139–154. doi:10.1007/s10750-014-1900-y

    CAS  Article  Google Scholar 

  17. Dattagupta S, Miles LL, Barnabei MS, Fisher CR (2006) The hydrocarbon seep tubeworm Lamellibrachia luymesi primarily eliminates sulfate and hydrogen ions across its roots to conserve energy and ensure sulfide supply. J Exp Biol 209:3795–3805. doi:10.1242/jeb.02413

    CAS  Article  PubMed  Google Scholar 

  18. Deweerdt S (2012) Comparative biology: looking for a master switch. Nature 492:S10–S11. doi:10.1038/492S10a

    CAS  Article  PubMed  Google Scholar 

  19. Drazen JC, Seibel BA (2007) Depth-related trends in metabolism of benthic and benthopelagic deep-sea fishes. Limnol Oceanogr 52:2306–2316. doi:10.4319/lo.2007.52.5.2306

    CAS  Article  Google Scholar 

  20. Fisher CR (1990) Chemoautotrophic and methanotrophic symbioses in marine invertebrates. Rev Aquat Sci 2:399–436

    CAS  Google Scholar 

  21. Ingram WC, Meyers SR, Brunner CA, Martens CS (2010) Late Pleistocene–Holocene sedimentation surrounding an active seafloor gas-hydrate and cold-seep fi eld on the Northern Gulf of Mexico Slope. Mar Geol 278:43–53. doi:10.1016/j.margeo.2010.09.002

    CAS  Article  Google Scholar 

  22. Kirkwood TB, Austad SN (2000) Why do we age? Nature 408:233–238. doi:10.1038/35041682

    CAS  Article  PubMed  Google Scholar 

  23. Kohyama T, Takada T (1998) Recruitment rates in forest plots: Gf estimates using growth rates and size distributions. J Ecol 86:633–639

    Article  Google Scholar 

  24. McClain CR, Allen AP, Tittensor DP, Rex MA (2012) Energetics of life on the deep seafloor. Proc Natl Acad Sci 109:15366–15371. doi:10.1073/pnas.1208976109

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  25. McCoy MW, Gillooly JF (2008) Predicting natural mortality rates of plants and animals. Ecol Lett 11:710–716. doi:10.1111/j.1461-0248.2008.01190.x

    Article  PubMed  Google Scholar 

  26. Paull CK, Jull AJT, Toolin LJ, Linick T (1985) Stable isotope evidence for chemosynthesis in an abyssal seep community. Nature 317:709–711. doi:10.1038/317709a0

    CAS  Article  Google Scholar 

  27. Ravaux J, Zbinden M, Voss-Foucart MF et al (2003) Comparative degradation rates of chitinous exoskeletons from deep-sea environments. Mar Biol 143:405–412. doi:10.1007/s00227-003-1086-8

    CAS  Article  Google Scholar 

  28. Ridgway ID, Richardson CA (2011) Arctica islandica: the longest lived non colonial animal known to science. Rev Fish Biol Fish 21:297–310. doi:10.1007/s11160-010-9171-9

    Article  Google Scholar 

  29. Roark EB, Guilderson TP, Dunbar RB et al (2009) Extreme longevity in proteinaceous deep-sea corals. Proc Natl Acad Sci U S A 106:5204–5208. doi:10.1073/pnas.0810875106

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  30. Roberts HH, Aharon P (1994) Hydrocarbon-derived carbonate buildups of the northern Gulf of Mexico continental slope: a review of submersible investigations. Geo-Marine Lett 14:135–148. doi:10.1007/BF01203725

    CAS  Article  Google Scholar 

  31. Robison B, Seibel B, Drazen J (2014) Deep-sea octopus (Graneledone boreopacifica) conducts the longest-known egg-brooding period of any animal. PLoS One 9:e103437. doi:10.1371/journal.pone.0103437

    Article  PubMed  PubMed Central  Google Scholar 

  32. SAS Institute Inc, Cary NC (1989-2013) JMP®, Version 11

  33. Williams GC (1957) Pleiotropy, natural selection, and the evolution of senescence. Evolution (N Y) 11:398–411

    Google Scholar 

Download references


This research was one part of a larger study led by Dr. Jim Brooks of TDI-Brooks that was jointly funded under the National Oceanographic Partnership Program (NOPP) by the US Bureau of Ocean Energy Management (BOEM), contract #0105CT39187, and the National Oceanic and Atmospheric Administration’s Office of Ocean Exploration and Research (NOAA OER). Many thanks to Erin Becker, Jeremy Potter, Liz Goehring, and Cindy Peterson for spending countless hours measuring tubeworms and to Stephanie Lessard-Pilon for her preliminary analysis of the tubeworm tag data. Collecting these tubeworms would not have been possible without the assistance of the captains and crew of the R/V Atlantis and NOAA Ship Ronald Brown and the crew and pilots of the DSV Alvin and ROV Jason II. We would also like to thank our anonymous reviewers for their valuable comments on improving this manuscript.

Author information



Corresponding author

Correspondence to Alanna Durkin.

Additional information

Communicated by: Sven Thatje

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Durkin, A., Fisher, C.R. & Cordes, E.E. Extreme longevity in a deep-sea vestimentiferan tubeworm and its implications for the evolution of life history strategies. Sci Nat 104, 63 (2017). https://doi.org/10.1007/s00114-017-1479-z

Download citation


  • Escarpia
  • Siboglinidae
  • Tubeworm
  • Cold seep
  • Longevity
  • Evolution