What otolith microchemistry and stable isotope analysis reveal and conceal about anguillid eel movements across salinity boundaries

Otolith microchemistry studies indicate that growth-phase (yellow stage) anguillid eels commonly shift at irregular intervals between fresh and saline waters, but this technique has not detected regular seasonal migrations across salinity zones. We tested the ability of otolith microchemistry and stable isotope analysis to detect migrations of American eels (Anguilla rostrata) between salinity boundaries in two small stream–estuary systems in Canada’s Bay of Fundy. Although the two methods showed concordant classifications of recent residence history, most eels caught in fresh water in spring (68.8–89.7 %) and fall (78.8–83.3 %) showed microchemical and isotopic signatures that reflected occupancy of saline waters. These eels were classified as migrants which had summered in saline waters and then migrated to freshwater wintering grounds where they retained their saline signatures. In summer, most eels (85.0–100.0 %) captured in fresh and saline water had recent microchemical and isotopic signatures matching the habitat of capture. Our results suggest that lifetime otolith microchemistry profiles are unable to detect eel wintering migrations, a failure that is likely due to winter depression of otolith accretion. Elucidation of seasonal eel movements requires cross-seasonal and cross-site sampling for the microchemistry and stable isotope methods, or tagging studies. Seasonal saline–fresh eel migrations may be more common than previously appreciated, underlining the need for conservation of both habitats, and connectivity between the two.     

Anguilla rostrata glass eel migration and recruitment in the estuary and Gulf of St Lawrence

This study describes catches of Anguilla rostrata glass eels and associated oceanographic conditions in the St Lawrence Estuary and Gulf. Ichthyoplankton survey data suggest that they enter the Gulf primarily in May, migrate at the surface at night, and disperse broadly once they have passed Cabot Strait. They arrive in estuaries beginning at about mid-June and through the month of July. Migration extends west up to Québec City, in the freshwater zone of the St Lawrence Estuary, 1000 km west of Cabot Strait. Anguilla rostrata glass eels travel between Cabot Strait and receiving estuaries at a straight-line ground speed of c. 10–15 km day⁻¹. Catches of fish per unit effort in estuaries in the St Lawrence system are much lower than those reported for the Atlantic coast of Canada. Low abundance of A. rostrata glass eels in the St Lawrence system may be due to cold surface temperatures during the migration period which decrease swimming capacity, long distances from the spawning ground to Cabot Strait and from Cabot Strait to the destination waters (especially the St Lawrence River), complex circulation patterns, and hypoxic conditions in bottom waters of the Laurentian Channel and the St Lawrence Estuary.     

Diverse migration strategy between freshwater and seawater habitats in the freshwater eel genus Anguilla

The freshwater eels of the genus Anguilla, which are catadromous, migrate between freshwater growth habitats and offshore spawning areas. A number of recent studies, however, found examples of the temperate species Anguilla anguilla, Anguilla rostrata, Anguilla japonica, Anguilla australis and Anguilla dieffenbachii that have never migrated into fresh water, spending their entire life history in the ocean. Furthermore, those studies found an intermediate type between marine and freshwater residents, which appear to frequently move between different environments during their growth phase. The discovery of marine and brackish-water residents Anguilla spp. suggests that they do not all have to be catadromous, and it calls into question the generalized classification of diadromous fishes. There has been little available information, however, concerning migration in tropical Anguilla spp. Anguilla marmorata, shows three fluctuation patterns: (1) continuous residence in fresh water, (2) continuous residence in brackish water and (3) residence in fresh water after recruitment, while returning to brackish water. Such migratory patterns were found in other tropical species, Anguilla bicolor bicolor and Anguilla bicolor pacifica. In A. b. bicolor collected in a coastal lagoon of Indonesia, two further patterns of habitat use were found: (1) constantly living in either brackish water or sea water with no freshwater life and (2) habitat shift from fresh water to brackish water or sea water. The wide range of environmental habitat use indicates that migratory behaviour of tropical Anguilla spp. is facultative among fresh, brackish and marine waters during their growth phases after recruitment to the coastal areas. Further, the migratory behaviours of tropical Anguilla spp. appear to differ in each habitat in response to inter and intra-specific competition. The results suggest that tropical Anguilla spp. have a flexible pattern of migration, with an ability to adapt to various habitats and salinities. The ability of anguillids to reside in environments of various salinities would be a common feature between tropical and temperate species without a latitudinal cline. Thus, the migration of Anguilla spp. into fresh water is clearly not an obligatory behaviour. This evidence of geographical variability among Anguilla spp. suggests that habitat use is determined by environmental conditions in each site.     

Variation in the sex ratio, size and age of longfinned eels within and among coastal catchments of south‐eastern Australia

Longfinned eels Anguilla reinhardtii were captured by both fishery‐dependent and independent sampling methods from three rivers in New South Wales, south‐eastern Australia. Sex ratios, catch per unit effort and population age and total length structure were examined in three zones (fresh water and upper and lower tidal) in the Hacking, Hawkesbury and Clarence Rivers. Females were found in relatively high proportions in all zones, ranging from 97% in a freshwater (non‐tidal) site down to 59% in a tidal site. Males were found primarily in tidal zones (only two of the 677 longfinned eels caught in non‐tidal fresh water were males), with the greatest proportions being found in the brackish upper tidal areas. The mean number of fish captured per trap was higher in the fresh water and upper tidal zones than in the lower tidal zones. The mean ±  s . e . age, 17·9 ± 0·3 years, and age range, 5–52 years for females were significantly higher than those of males 12·2 ± 0·4 years; range 5–22 years, which is typical of other anguillid species. Longfinned eels captured in fresh water were found be significantly larger and older than those in tidal zones due to the almost exclusive predominance of females.     

Validation of daily otolith increments in glass-phase American eels Anguilla rostrata (Lesueur) during estuarine residency

Prior to making inferences from otoliths about the residence time and growth rate of glass-phase anguillid eels Anguilla in estuaries, it is necessary to validate the deposition rate of microincrements in the otoliths. Glass-phase American eels Anguilla rostrata (Lesueur), which had been captured near the mouth of an estuary in Maine, USA, prior to freshwater exposure, deposited increments at a daily rate at ambient temperature and salinity in a field and laboratory study. The regression for glass eels not possessing a transition ring was: I=0.976(D-1)+0.434, where I is the number of otolith increments distal to a fluorescent mark placed on the otolith at the beginning of the experiment, and D is the number of days in the experiment, which ranged from 7 to 49. The slope was not significantly different than 1. Unexpectedly, many glass eels deposited the transition ring during the experiment, although this ring had previously been thought to mark entry into fresh water. The regression for these glass eels was: I=0.961(D-1)-3.880, and the slope was not significantly different than 1. The negative intercept suggests that approximately 4 days were lost from the otolith record during deposition of the ring. This study demonstrated daily deposition of increments prior to freshwater exposure and demonstrated that deposition of the transition ring is not linked to freshwater entry.     

Migration and habitat use of the tropical eels Anguilla marmorata and A. bicolor pacifica in Vietnam

The patterns of use of marine and freshwater habitats by the tropical anguillid eels Anguilla marmorata and A. bicolor pacifica were examined by analysing the otolith strontium (Sr) and calcium (Ca) concentrations of yellow (immature) and silver (mature) stage eels collected in Vietnamese waters. In A. marmorata, the change in the Sr:Ca ratios outside the high Sr:Ca core was generally divided into three patterns: (1) typical catadromous life history pattern; (2) constant residence in brackish water; and (3) habitat shifting between sea and brackish waters with no freshwater life. In A. bicolor pacifica, no eels had a general life history as freshwater residents. The eels were also divided into three patterns: (1) constant residence in sea water; (2) constantly living in brackish water; and (3) habitat shifting from brackish to sea water with no freshwater residence. The mean Sr:Ca ratio value after recruitment to coastal waters ranged from 1.73 to 5.67 × 10^?3 (mean 3.2 × 10^?3) in A. marmorata and from 2.53 to 6.32 × 10^?3 (mean 4.3 × 10^?3) in A. bicolor pacifica. The wide range of otolith Sr:Ca ratios in both species indicated that the habitat use of these tropical eels was facultative among fresh, brackish, and marine waters during their growth phases after recruitment to coastal areas. Tropical eel species may have the same behavioural plasticity as temperate anguillid species regarding whether to enter freshwater or to remain in estuarine and marine environments.     

Differential effects of origin and salinity rearing conditions on growth of glass eels of the American eel Anguilla rostrata: implications for stocking programmes

In this study, growth patterns were monitored in controlled fresh and brackish water (BW) conditions for 7 months during Anguilla rostrata glass eel and elver stages. Null hypotheses tested were that there is no significant difference in growth between glass eels (1) collected from two geographic regions typified by different sex ratios, (2) reared in fresh and BW and (3) due to origin × salinity interactions. It was found that young A. rostrata from Mira River (MR, Cape Breton, Nova Scotia, Canada, an area where both males and females occur) grew faster than those from Grande-Rivière-Blanche (Québec, Canada, an area where population are highly skewed towards females; 99–100%). Anguilla rostrata from both origins also grew faster in BW, although there was a trend for origin × salinity interactions whereby this effect was more pronounced for fish from the MR. The results support the hypothesis that salinity can influence growth patterns, as possibly can quantitative genetic differences between A. rostrata glass eels from different origins. Possible explanations for these patterns and potential consequences for sex determination and translocation programmes are discussed.     

Regional variation and the effect of lake: river area on sex distribution of American eels

Silver phase American eels, Anguilla rostrata , were collected while migrating from five rivers in Maine, U.S.A. Sex ratios varied from 49 to 98% male for these rivers and had a range of 46% over a 30 km distance between the mouths of three rivers. The proportion of male eels was inversely related to the amount of lacustrine habitat in the five drainage areas ( r =−0·95, P =0·014). A combination of these sex ratios and published data from two Nova Scotia rivers showed large variation in the proportion of male eels within 1° of latitude. Thus, the hypothesis from the literature that the distribution of the sexes is dependent upon distance of larval transport was not supported. Eels migrating from lacustrine habitats within a river were predominately female, while eels migrating from fluvial habitats were predominately male, regardless of upstream distance. Apparently river habitat influences the distribution of the sexes and may play a role in sex determination.     

Migratory patterns and contribution of stocking to the population of European eel in Lithuanian waters as indicated by otolith Sr:Ca ratios

Otolith Sr:Ca ratios were examined to evaluate the contribution of the stocked eel Anguilla anguilla elvers, which have been stocked in Lithuanian waters and mixed with naturally recruited eels for several decades, to the native eel population. Stocked eels were identified by the freshwater signature (Sr:Ca ratios <2·24 × 10⁻³) on the otolith after the glass eel stage. Naturally recruited eels, that had migrated through the North and Baltic Seas, were characterized by an extended seawater and brackish-water signature (Sr:Ca ratios >3·23 × 10⁻³) after the glass eel stage. Of 108 eels analysed, 21 eels had otolith Sr:Ca ratio profiles consistent with stocking while 87 showed patterns of natural recruitment. The ages of naturally recruited eels arriving in Lithuanian fresh waters varied from 1 to 10 years, with a mean ±  s.d . age of 5·2 ± 2·1 years. Eels from the inland Lake Baluošai were all freshwater residents of stocked origin. Stocked eels, however, accounted for only 20% of the eels from the Curonian Lagoon and 2% of eels sampled in Baltic coastal waters. This finding does not support the hypothesis that the eel fishery in the Curonian Lagoon depends mostly on stocking.     

The exotic American eel in Taiwan: ecological implications

Six American eels Anguilla rostrata were identified in the Kaoping River of south-western Taiwan by sequencing the mitochondrial cytochrome b gene ( mtCyt-b ). This was the first record of an exotic eel species in the natural waters of Taiwan. The exotic eels may have been imported from North America at the elver stage for aquaculture and escaped from culture ponds.     

Inter and intra-estuary variability in ingress, condition and settlement of the American eel Anguilla rostrata: implications for estimating and understanding recruitment

The objective of this study was to quantify spatial and temporal variability of anguillid glass eel ingress within and between adjacent watersheds in order to help illuminate the mechanisms moderating annual recruitment. Because single fixed locations are often used to assess annual recruitment, the intra-annual dynamics of ingress across multiple sites often remains unresolved. To address this question, plankton nets and eel collectors were deployed weekly to synoptically quantify early stage Anguilla rostrata abundance at 12 sites across two New Jersey estuaries over an ingress season. Numbers of early-stage glass eels collected at the inlet mouths were moderately variable within and between estuaries over time and showed evidence for weak lunar phase and water temperature correlations. The relative condition of glass eels, although highly variable, declined significantly over the ingress season and indicated a tendency for lower condition A. rostrata to colonize sites in the lower estuary. Accumulations of glass eels and early-stage elvers retrieved from collectors (one to >1500 A. rostrata per collector) at lower estuary sites were highly variable over time, producing only weak correlations between estuaries. By way of contrast, development into late-stage elvers, coupled with the large-scale colonization of up-river sites, was highly synchronized between and within estuaries and contingent on water temperatures reaching c. 10−12° C. Averaged over the ingress season, abundance estimates were remarkably consistent between paired sites across estuaries, indicating a low degree of interestuary variability. Within an estuary, however, abundance estimates varied considerably depending on location. These results and methodology have important implications for the planning and interpretation of early-stage anguillid eel surveys as well as the understanding of the dynamic nature of ingress and the spatial scales over which recruitment varies.     

Discrepancies between otoliths of larvae and juveniles of the American eel: is something fishy happening at metamorphosis?

The use of otoliths to interpret early life history in fishes depends upon the assumptions that otoliths record past events accurately and consistently and that records of events in otoliths are continuous. Both the number of growth microincrements ( I ), and the radii ( R ,μm) of otoliths of American eel Anguilla rostrata , leptocephali increased linearly and highly significantly with leptocephalus body length ( L , mm), as expected on the above assumptions ( I , =2·29 L , − 5·75 and R , =1·05 L , + 12·02, r ²,=0·938 and 0·931, n , =20). In contrast, the number of increments and the radii of the leptocephalus growth zones of otoliths of glass-phase American eels were not related to body length, and they were lower than predicted by the relationships developed for leptocephali. Thus, otoliths of American eels apparently violate one or both assumptions. Possibly, the margin of the otolith is resorbed during metamorphosis from leptocephalus to glass eel, perhaps as part of calcium metabolism as skeletal elements are being formed.     

Determination of the age of young American eels, Anguilla rostrata, in fresh water, based on otolith surface area and microstructure

The time elvers of the American eel, Anguilla rostrata , spend in an estuary prior to their migration into fresh water was assessed. A distinct mark was formed on elvers' otoliths during their first 2 to 3 weeks in the river estuary. This mark was used to distinguish between growth in fresh water and in salt water. Migrating eels collected at a falls 4 km from the estuary exhibited bimodal length and weight distributions. Frequency distributions showed that eels collected in the estuary were smaller and had smaller otoliths than eels collected at the falls, indicating that elvers do not reach the falls in the same year as they enter the estuary. The three modal groups most probably represent three age–classes. However, the otoliths of elvers collected in the estuary had only the mark of transition whereas eels in the first and second mode at the falls usually had two rings (1–4) and four rings (3–6) per otolith, respectively, in addition to the mark of transition, as viewed under SEM. The possibility that ring formation is not annual means that the use of otoliths for the age determination of eels in this study has not been validated.     

Advanced sexual maturation before marine migration of Anguilla bicolor bicolor and Anguilla marmorata at Réunion Island

Maturing sub-adults of two species of anguillid eels (a female Anguilla bicolor bicolor and a male Anguilla marmorata ) were collected for the first time at Réunion Island, western Indian Ocean. Both were silver eels, i.e. maturing eels at the onset of their spawning migration, and characterized by advanced sexual maturation that has been only observed in Anguilla dieffenbachii from New Zealand.     

Effect of lunar periodicity on the locomotor activity of silver-stage Japanese eel,Anguilla japonica

Lunar periodicity has been thought to influence the onset of the spawning migration of anguillid eels. In this study, we measured daily locomotor activity of 8 silver-phase Japanese eels Anguilla japonica in outdoor tanks to examine the effect of lunar periodicity on their activity and the following seaward migration. The activity of silver eels was highest around the new moon during the early part of the experiment, which is the ordinary season of seaward migration in Japan. The observed patterns of activity may reflect the importance of the lunar cycle for the onset of the spawning migration in anguillid eels.     

Anguillicola crassus infection in Anguilla rostrata from small tributaries of the Hudson River watershed, New York, USA

We studied the invasion of the exotic nematode parasite Anguillicola crassus in the American eel Anguilla rostrata using tributaries of the Hudson River estuary. Yellow-phase American eels were sampled from 6 tributaries, and their swim bladders were examined for nematode infection. Prevalence averaged 39% with an intensity of 2.4 nematodes per eel. Parasite distribution was not significant along a latitudinal gradient; on the other hand, physical barriers (dams and natural waterfalls) significantly reduced infections upstream. Urbanization may increase the susceptibility of eels to infection; we found significantly elevated infection rates when urbanized lands exceeded 15% of the tributary catchment area. Yellow-phase eel condition was not affected by parasite infection. The invasion of the entire Hudson River watershed is ongoing and therefore will continue to be a management concern. Further analysis of the parasite-host interaction in North America is warranted.     

Discrimination of New Zealand stream waters by glass eels of Anguilla australis and Anguilla dieffenbachii

This study tested the hypotheses, using glass eels of longfin eels Anguilla dieffenbachii and shortfin eels Anguilla australis migrating into fresh waters, (1) that both species prefer water from their river of collection to well water, (2) that shortfin eels prefer lowland, pastoral stream water to mainstem river water, (3) that longfin eels are attracted to both waters but do not prefer one to the other, and (4) both prefer water scented with geosmin, a widely occurring metabolite of bacteria and algae, to well water. Glass eels of both species from a river on the west coast of South Island, New Zealand, and shortfin glass eels from an east coast river significantly preferred water from their river of capture to well water. Two to three times as many eels chose their own river water as chose well water. Longfin eels were rare in the east coast river. Shortfin glass eels from the two rivers chose lowland stream water to mainstem river water about two to one in three experiments with different pairs of waters to which they had no prior exposure. Longfin glass eels significantly chose mainstem river water over lowland water in one pair but showed no preference when presented with a different pair. Reactions to solutions of geosmin at concentrations of 10^–5-10^–7 mg 1^–1 were inconclusive, with geosmin being preferred significantly, by shortfin eels, in only one experiment. The interspecific differences in discrimination of natural waters demonstrated in this study, with shortfin eels preferring lowland waters and longfin eels more indifferent to water types, are in broad agreement with both the distribution of adults and observations on their habitat preferences.     

Development of a key using morphological characters to distinguish south-western Indian Ocean anguillid glass eels

Two morphological characters were examined in 4099 anguillid glass eels sampled in four south-western Indian Ocean islands, and resulting identifications were tested using genetic analysis. Distance between the origin of the dorsal and anal fins as related to total length and tail and caudal fin pigmentation enabled formulation of a useful field identification key, which was able to discriminate 4036 glass eels of the regional species Anguilla bicolor bicolor , Anguilla marmorata and Anguilla mossambica , and excluded the occurrence of Anguilla nebulosa labiata in the sample.     

Differences in size and growth rates of male and female migrating Japanese eels in Pearl River, China

The Sr/Ca ratios in otoliths of silver Japanese eels Anguilla japonica , in Pearl River, China, indicated that both sexes did not stay in brackish water and grew in fresh water from the glass eel stage until spawning migration. This did not support the hypothesis that females tended to distribute upstream and males might be restricted to estuaries. The back-calculated total length of males at glass eel stage was not significantly different from that of females, indicating that the hypothesis that small glass eels became males and larger ones became females may not be true. The mean (±S.D.) age and total length of males at migration were 6·4±1·6 years and 48·3±4·5 cm, which were significantly smaller than for females, 8·3±1·6 years and 61·4±4·1 cm. The age of migration was related inversely to growth rate for both sexes. Growth parameters of the von Bertalanffy growth equation were K =0·21 cm year°1, L _∞=55·7 cm and t _o=-0·55 year for males and K =0·14 cm year⁻¹, L _∞=77·5 cm and t _o=-0·60 year for females. The difference in asymptotic length ( L _∞) between males and females may be because females postpone migration to achieve larger size for maximizing reproductive success.     

Influence of different rations and water temperatures on the growth rates of shortfinned eels and longfinned eels

Juvenile (12–152 g) shortfinned eels Anguilla australis and longfinned eels A. dieffenbachia caught in New Zealand streams were fed squid mantle Nototodarus spp. 4 days per week in laboratory experiments. A linear multiple regression equation showed the amount of food eaten (0–2·7% w day⁻¹) explained 77·7% of the variation in specific growth rates (–0·60 to +1·07% w day⁻¹) among individual eels, while previous growth rates, water temperature (10·0–20·6°C), and eel weight (12–152 g) explained a further 5·6, 1·4 and 0·8%, respectively. Growth in length ranged from –0·3 to +0·9 mm day⁻¹. Eels which were starved and then given high rations grew substantially faster than expected. Once growth rates were adjusted for differences in ration and other factors, there were no significant differences in growth rates between species or individual fish. Growth of shortfinned eels fed maximum rations of commercial eel food depended on fish size and water temperatures and ceased below 9·0°C. Growth rates in the wild were substantially less than the maximum possible, after seasonal changes in water temperatures were taken into account, indicating that food supplies and not low water temperatures were controlling growth rates in the wild.