One life, in ten parts
The life of a London eel: from the Sargasso Sea to the River Wandle, and back
There are eels in the Wandle, the Roding, the Lea and the Thames at Battersea. Every one of them hatched in the Atlantic, thousands of kilometres from London, in a place where nobody has ever seen an eel spawn. This is one such life. Scroll, and it grows up.
One representative eel, pieced together from tagging studies, fish surveys and the traps London volunteers check each summer. No single animal has been followed through all of it; the steps say what was measured, and where.
Scroll to begin
I
Nobody has seen this
It starts in warm water south and south-east of Bermuda, somewhere in a stretch of ocean about 2,000 km wide and about 5,700 km from the Thames in a straight line.
No one has seen it happen. No spawning European eel, and no European eel egg, has ever been found at sea.
What has been found is larvae. In March and April 2014 two research ships netted 408 of them here, the smallest 5.5 mm long.
The smallest larvae turn up in this sea. On that the case for it as the eel’s birthplace was built, in the 1920s, and it still stands without a single egg to prove it. (The 3 mm hatchling is known only from eels bred in laboratories.)
The gap is old. Aristotle, who never found an eel carrying eggs or milt, concluded that eels “proceed neither from pairing nor from an egg”.
In 1876 a medical student working at the zoological station in Trieste examined about 400 eels to check a claim that a certain lobed organ was the testis. He could neither confirm it nor rule it out. His name was Sigmund Freud.
II
A leaf in the current
The hatchling becomes a leptocephalus: a transparent, leaf-shaped larva, flattened from side to side. European eel larvae can reach almost 9 cm.
What they eat is still argued over. DNA from their guts points to jelly-like plankton; a leading review says marine snow, the drifting debris of the upper ocean.
Then it crosses the Atlantic. How long that takes is not settled. Growth rings in ear bones once suggested six months. Drift models say a larva carried by the currents needs at least ten, and one puts the average at 21 months.
The same model lets fewer than one larva in 500 survive the crossing. That is an estimate from a model; nobody has counted.
It cannot aim for London. European eels are a single population that breeds at random: a whole-genome study found no geographic pattern in the genes of eels from across Europe and North Africa.
Where a larva makes landfall appears to be largely chance. Its parents may have grown up anywhere in that range. This one arrives off the Thames.
III
Glass
At the edge of the continental shelf the leaf remakes itself into a glass eel: a transparent thread, about 7 cm long on average when it reaches Europe’s coasts.
It has no colour yet. It also has no sex. In eels that is not fixed by genes; it is thought to be settled later by how fast, and how crowded, this one grows. Until then, this page says “it”.
IV
Up the Thames on the tide
Ahead lie 99.4 km of tidal river, from Southend Pier to Teddington. Twice a day the sea pushes up it: at London Bridge the water rises and falls 6.6 m on a spring tide.
On the 2026 tide tables the flood at London Bridge runs for about six hours and the ebb for about six and a half.
Glass eels in tidal estuaries are described as doing something neat: rising into the water when the tide floods, to be carried upriver, and keeping to the bottom when it ebbs.
The classic measurement is from American eels, in a river in Maine in 1980–81. On most nights they gained between 0.8 and 4.2 km a tide. Keep scrolling and watch the water.
For the Thames there is a computer model rather than a net. A 2021 study sent simulated young eels up the estuary and compared the result with real catches in an eel trap at Brentford.
Its authors report that holding to the riverbed on the ebb was essential for the model to match the trap.
It does not rush. A study of the Thames in the 1980s found that most new arrivals seemed to spend at least a year in the estuary before moving into fresh water.
By then they have turned dark. A pigmented young eel is called an elver.
Just upstream of Tower Bridge it passes the old Billingsgate fish market. Around 1850 the market’s salesmen told Henry Mayhew that 9,797,760 eels were sold there in a year. Most came alive from Holland, in boats moored off the quay.
Dutch eel boats were still being photographed there in 1935–36.
By the 1950s pollution had closed this reach to almost every fish that came up from the sea. In a survey of London’s fishes published in 1958, Alwyne Wheeler listed the migrants that had “ceased to exist in the Thames”, among them salmon and flounder.
And then: “The only fish which ascends the Thames from the sea to-day is the eel.”
At Wandsworth, 80.2 km from Southend Pier, a small river comes in on the south bank: the Wandle. It turns in.
V
The wall and the count
Above the tide, London’s rivers are staircases of weirs and sluices. A 2012 report quotes an Environment Agency count of 2,393 barriers in the Thames catchment.
A small eel can creep up a wet wall. A French study found that only those under 10 cm manage it, and that few of them get over.
So people build ramps. A 2019 paper counted 102 eel passes built since 2005 across the Thames river basin district, which also takes in the Medway: slopes lined with something wet to climb.
Beside some of them sits a trap. From April to September volunteers trained by the Zoological Society of London lift the lid, count the young eels, measure them and let them go upstream. More than 1,300 people have been trained since 2011.
This is what they count. At Molesey Weir, a little upstream of the tidal Thames, the traps caught 14.63 elvers for every day of trapping in 2013, the best year in a record that starts in 2012.
In 2025 they caught 0.22: on average, one elver for every four and a half days of trapping.
That year the trap on the Roding at Redbridge recorded 0.25 a day, its lowest since 2012. Trap counts swing sharply: 2013 was an unusually good year at Molesey, and the Redbridge weir was altered in 2024. The longer, coast-wide record shows the direction more plainly.
On the coasts of the North Sea, the index of glass eels arriving stood in 2024 at 1.3% of its 1960–1979 average, by the reckoning of ICES, the scientific body that advises governments on fish stocks. Its advice for 2026: catch none.
Not every glass eel that reaches Europe stays. On 15 February 2017 Border Force officers at Heathrow seized 200 kg of live glass eels bound for Hong Kong, valued at £5.7 million or more on the black market. At 2,200 to 3,800 to the kilo, that is several hundred thousand animals.
The trader, from Chessington, was given a suspended sentence. The eels were sent back to Spain for release.
VI
The long wait
It settles. The Environment Agency’s records hold 1,734 fish surveys inside Greater London since 1978. Eels turned up in 739, 8,588 of them in all.
The Wandle accounts for 4,086, caught between 1988 and 2025. These are survey catches added up, not a population. On 7 October 2025 a single survey at Mapleton Road counted 238.
Now it is a yellow eel, and its days are quiet. Tagged eels in a Dorset estuary moved mostly between sunset and sunrise, within a range of 630 m to about 4 km.
In estuaries on the French side of the Channel their guts held small crustaceans, shore crabs, young flounder and smaller eels. In water colder than 8–10 °C, eels stop feeding.
It grows slowly: about 3 cm a year in one Norwegian river. The eels measured in London’s freshwater surveys have a median length of 357 mm. The longest, caught in the pool below Molesey Weir in 2014, was 1.1 m.
Past half a metre, a biologist would write it down as a female. That is a working rule, based on length alone: one recent study tentatively classes eels over 50 cm as female and those under 46 cm as male, calls anything from 40 to 50 cm unpredictable, and notes exceptions both ways. When an eel’s sex is actually settled cannot be seen from outside.
This representative eel is female. From here on, the page says she.
How long she stays is the widest number in her life. One review gives 5 to 25 years. Silver females leaving one Norwegian river averaged about 20, and ranged from 5 to 39.
Londoners have eaten eels like her for centuries. In 1086 the fishery at Petersham owed Chertsey Abbey 1,000 eels a year. Around 1850 Mayhew reckoned there might be 500 sellers of hot eels and pea soup in the streets on a Saturday.
London had over 100 pie-and-mash shops at the end of the 19th century, 87 in 1995 and about 30 in 2024. Many of London’s eels now come from Northern Ireland, by the London Museum’s account.
VII
The change
One summer she begins to change. Her eyes enlarge and fill with rods, the cells that work in dim light, while the cones thin out. Her pectoral fins lengthen. Her belly pales.
She is now a silver eel.
Her gut shrinks and loses its ability to absorb food. Silver eels are thought not to feed again, though one Swedish study suggests some may, early on.
Whatever fat she has laid down is the fuel for what comes next.
And she leaves unfinished. A silver eel sets out with her gonads at less than 2% of her body weight, not yet able to breed.
Physiologists argue that this is less a metamorphosis than the start of puberty. The rest of it happens somewhere in the ocean.
VIII
The night she goes
Most silver eels leave Europe’s rivers between August and December. On an Irish and a Norwegian river, each watched for more than forty years, more of them moved on nights of high and rising water, and fewer under a strong moon.
So: a wet, dark night. Down the Wandle, out at Wandsworth, and back along the 80.2 km of tideway she once climbed.
Nobody counts her out. The official plan for the Thames’s eels says there is “no direct measurement” of how many silver eels escape to sea, and no study that tracked a Thames eel into the ocean could be found for this page.
IX
The road back
What is known comes from eels tagged elsewhere. European teams fitted 707 silver eels with tags. Data came back from 206, and 87 of those had got clear of the coast.
Those released into the North Sea from Germany were followed for about a month. Four turned south towards the English Channel. None had reached the open ocean when the record stopped.
Across all the tracks, speeds ran from 3 to 47 km a day. In the ocean the eels swam deep by day and came up at night, mostly between 200 and 1,000 m.
The tag records suggest that about half of the eels followed for more than a week were eaten, some near the coast and some in the open sea; the tags may themselves have made them easier prey. The routes bent towards the Azores, and no eel released from mainland Europe has been followed further than that.
The nearest thing to proof came in 2022. Of 26 eels tagged in the Azores, 21 returned usable records, and one tag surfaced inside the presumed spawning area.
At the speeds measured, many eels may need more than a year. A laboratory test suggests the fuel can last: nine young farmed eels, not yet silver, swam an average of 5,533 km in a freshwater tunnel over about six months, and the authors concluded that an eel’s reserves are “in principle, sufficient”. A tunnel is not an ocean.
X
No one sees the end
If she arrives, and if her reserves have lasted, she spawns. Eels of her kind carry developing eggs in the millions: in one German river, counts in females setting out ran from 1.44 to 7.46 million.
Then she is believed to die. Neither the spawning nor the death has ever been observed in a European eel in the wild.
Her young will not look for the Wandle. Genetic studies find no sign that eels return to the river their parents left.
A larva of hers may make landfall anywhere from the Baltic to North Africa, and some stranger’s will come up the Thames.
Each April the volunteers go back to the weirs, lift the lids and count. In 2025, at Molesey, it was one elver for every four and a half days of trapping.
Whatever is in the trap hatched in a sea where no one has seen an eel spawn, from parents no one has found, and came to London by chance. When you next cross the Wandle, the Roding, the Lea or the Thames at Battersea after dark, something like her may be underneath, waiting.
What is measured, and what is imagined
The eel on this page is a composite. No animal has been followed from the Sargasso Sea to a London river and back, and the page never claims one has. Each step rests on a study or a record, named in the sources below, and most say where the evidence comes from when it is not the Thames: Maine for the tidal behaviour, Dorset for the night-time range, Norway for ages, Norway and Ireland for the nights of departure, France and the Netherlands for the changes of silvering, Germany and the Azores for the tagged migrants.
The straight-line distance is computed here, from the outer Thames estuary (51.50° N, 0.75° E) to the box in which recent papers draw the presumed spawning area (24–31° N, 50–70° W): 5,656 km to its centre, 4,694 km to its nearest corner and 6,629 km to its farthest. Real routes are longer and have not been measured for any Thames eel.
River distances are computed from the Port of London Authority’s table of distances, re-based here to Southend Pier Head. The flood and ebb durations are averaged from every high and low water predicted for London Bridge in 2026. The picture of the tide compresses a year or more of tides into six.
The trap figures are elvers (eels under 120 mm) per day of successful trapping, from the 2025 report of ZSL’s Thames European Eel Project. Molesey Weir is 8.1 km above the tidal limit at Teddington; a Wandle eel would not pass it. The trap is a nearby record of young eels moving into the freshwater Thames, not a count of arrivals in the Wandle. The figures vary a great deal from year to year and from site to site. The older claim of a decline of “over 99%” applies to one tributary, the Darent, in 2005–09 against the mid-1980s.
The Environment Agency figures are catches in fish surveys made with different methods and effort between 1978 and 2026. They show where eels have been recorded, not how many there are.
The drawings are not portraits and the scenes are schematic. The scale bar beside the eel is true to the length given at each step.
Elvers per day of trapping at ZSL’s three index sites, 2012–2025
| Year | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Medway – Allington Lock | 10.90 | 133.30 | 66.68 | 2.34 | 0.48 | 1.49 | 0.76 | 0.87 | – | 3.29 | 1.40 | 2.21 | 0.60 | 3.84 |
| Roding – Redbridge Roundabout | 0.08 | 0.47 | 7.20 | 2.36 | 0.83 | 1.15 | 3.61 | 1.79 | – | – | 0.73 | 0.32 | 2.00 | 0.25 |
| Thames – Molesey Weir | 0.82 | 14.63 | 2.10 | 1.68 | 1.63 | 0.82 | 0.67 | 0.57 | 0.61 | 0.90 | 0.09 | 0.04 | 0.23 | 0.22 |
A dash means no monitoring that year (Covid-19). Only the Roding site is clearly inside Greater London; Allington Lock is in Kent.
Sources
- ZSL, Thames European Eel Project Report 2025 (trap counts, volunteers)
- Pecorelli et al. 2019, Freshwater Science 38:281–291 (eel passes, monitoring sites)
- ZSL, Citizen Science European Eel Monitoring Report 2012 (2,393 barriers)
- ICES Advice 2025, European eel, Table 8 (recruitment indices; zero-catch advice)
- Environment Agency, Ecology & Fish Data Explorer (fish surveys in Greater London, OGL v3)
- Port of London Authority, Tide Booklet 2026 (distances, tidal levels, predictions)
- Thames Eel Action Plan 2025 (no direct measurement of silver-eel escapement)
- Wheeler 1958, The Fishes of the London Area, London Naturalist 37:80–101
- Naismith & Knights 1988, J. Fish Biology 33:161–175 (Thames elver runs; abstract)
- Benson et al. 2021, Ecological Modelling 443:109448 (Thames tidal-transport model; abstract)
- McCleave & Kleckner 1982, ICES J. Mar. Sci. 40:262–271 (tidal transport, American eel; abstract)
- Wright et al. 2022, Scientific Reports 12:15362 (Azores tags; no eggs or spawning adults found)
- Hanel, Marohn & Westerberg 2022, Scientific Reports 12:11778
- Miller et al. 2019, Biology Letters 15:20180835 (408 larvae across 2,000 km)
- Righton et al. 2016, Science Advances 2:e1501694 (707 tagged eels; hatch size)
- Bonhommeau et al. 2009, Fisheries Oceanography 18:371–385 (drift model; abstract)
- Miller et al. 2015, Biological Reviews 90:1035–1064 (larval size; abstract)
- Ayala et al. 2018, Scientific Reports 8:6156 (larval diet)
- Enbody et al. 2021, PNAS 118:e2022620118 (one population)
- Cresci 2020, Biological Reviews 95:1273–1286 (glass eels; 5–25 years; abstract)
- Durif et al. 2020, Ecology and Evolution 10:4801–4815 (glass-eel size, ages, growth, feeding temperature)
- Williamson et al. 2026, J. Fish Biology 108:1943–1959 (sex determination; length rule)
- Legault 1988, Bull. Fr. Pêche Piscic. 308:1–10 (climbing; abstract)
- Walker, Godard & Davison 2014, Aquatic Conservation 24:155–165 (Dorset; abstract)
- Denis et al. 2022, PLoS ONE 17:e0270348 (diet; abstract)
- Palstra & van den Thillart 2010, Fish Physiol. Biochem. 36:297–322 (silvering; gonad mass; swim trial)
- Pankhurst 1982, J. Fish Biology 21:127–140 (eyes; abstract)
- Righton et al. 2021, Fish and Fisheries 22:762–788 (silver eels and feeding)
- Svedäng & Wickström 1997, J. Fish Biology 50:475–486 (abstract)
- Aroua et al. 2005, Neuroendocrinology 82:221–232 (silvering as puberty; abstract)
- Sandlund et al. 2017, Ecology and Evolution 7:5956–5966 (nights of departure)
- van Ginneken et al. 2005, J. Experimental Biology 208:1329–1335 (swim tunnel; abstract)
- van Ginneken et al. 2007, Animal Biology 57:63–77 (silvering over summer; abstract)
- Gollock, Curnick & Debney 2011, Hydrobiologia 672:33–37 (Darent, over 99%; abstract)
- London Museum: jellied eels (Northern Ireland)
- London Museum: London’s pie and mash shops (shop counts)
- Tsukamoto & Miller 2021, Fisheries Science 87:11–29 (larval diet review; abstract)
- Bonhommeau et al. 2009, J. Fish Biology 74:1891–1914 (drift model with mortality; abstract)
- Durif, Dufour & Elie 2005, J. Fish Biology 66:1025–1043 (silvering stages; abstract)
- Reismann & Frankowski 2022, J. Fish Biology 100:320–324 (oocyte counts; abstract)
- Aristotle, History of Animals, Book VI, trans. D’Arcy Thompson
- Freud 1877, Sitzungsberichte der Kaiserlichen Akademie der Wissenschaften 75
- Mayhew, London Labour and the London Poor, vol. 1, pp. 63, 160
- London Museum: Dutch eel boats moored off Billingsgate, 1935–36
- Open Domesday: Petersham
- Cefas, 6 March 2020 (Heathrow seizure and sentence)
- UNODC, World Wildlife Crime Report 2020, ch. 7 (glass eels per kilogram)