Otters have returned to the Thames and Cherwell valleys in recent years. They’re harder to spot than badgers, because they don’t raid dustbins. When someone recently posted a video of an otter she had seen swimming near Magdalen Bridge, after the Boxing Day floods, one viewer exclaimed ‘Amazing. I have been waiting years to see an otter – and I live on a boat.’

If not in the flesh, you can see local otters on a webcam, because there are some who live by Magdalen meadows, watched by college webcams, some sequences from which have been posted online.  LMH has also posted otter footage. They seem to be reported on the relatively quiet Cherwell more than on the Thames,

That otter numbers are increasing seems to be agreed, with a reduction of pesticides in water generally credited. The fifth national otter survey, 2009-10, told a story of change.

 In the upper Thames region, revival is dated from the 1990s. The Oxford Mail  reported their return in 2010. The Thames Valley environmental records centre illustrates their return with maps of spottings across the two counties of Oxfordshire and Berkshire.

Our otters are Eurasian otters. They mainly eat fish. They have fishing territories which they defend. They are solitary, and get together only to mate. Though cubs hang around with their mother for more than a year, learning the ropes. ‘A male otter may use more than 20km of river/stream.’ So are there several otters on the Cherwell, or the same otter(s) sighted in several places?

Bernard Landgraf, CC BY-SA 3.0, via Wikimedia Commons

Their activities may concern anglers (or fishkeepers) In 2018 a petition to Parliament suggested that numbers had bounced back to the point where non-lethal measures to control them were needed. The petition attracted 12K signatures, esp in counties north and east of Oxford, but less so in Oxford and Oxford East. The government declined to act.

Anyone spotting one is encouraged to report the sighting to the Environmental Records Centre.


As it approaches the old city of Oxford from the north, just before it receives the Peasmoor brook from the east, the Cherwell divides. One part loops eastwards, past the Peasmoor brook entry and Parson’s Pleasure.  A gravel bank or ait within this braid has been reinforced to serve as a continuous footpath: Mesopotamia. Another parts cuts through to the west, to rejoin the main branch around Magdalen bridge. This western braid is the Holywell Mill Stream. Is it a channel dug as a mill stream?

The stream divides Merton’s Music Meadow from the New College Recreation Ground, then runs under Manor Road as it heads towards St Catherine’s. Just before the stream reaches Magdalen’s north wall is a Magdalen graduate house, Holywell Ford, which stands across it, where the mill once stood. The stream then passes through Magdalen grounds, with first the Grove, then college buildings on one bank, the Water Meadow on the other. It then rejoins two other braids of the Cherwell either side of Magdalen Bridge.

The parish of Holywell sits across these rivers. It’s shaped like a short-toed boot, ready to kick eastwards the buildings along the Marston Road. The back of the boot is Parks Road. At the heel, the boot runs along the old city wall, from Catte St through New College to the bastion where Longwall street meets St Cross st. So Holywell st, the street that leads to the holy well, lies wholly within the parish. The main stretch of the sole then runs along the northern edge of Magdalen grounds: Holywell Ford (not originally but now owned by Magdalen) lies just inside the parish. The sole extends to just before St Clement’s church, then the toe curves up along the main course of the Cherwell, Mesopotamia Walk lying just within the boundary. The top of the boot runs along the top of the Parks, paralleling Norham Road to the south.

Holywell parish, from 2nd series OS map 1889-1914 (

Holywell was also a manor, and as rebuilt in the sixteenth century, the manor house survives, though much altered, as Holywell Manor, a Balliol graduate house. But Balliol only lease the house, which, like much of the land round here, is, or was once, owned by Merton. The Parks were originally Holywell Manor’s fields, developed by Merton as a park, then sold to the University in the mid nineteenth century, to serve as University parks and recreation ground. Merton’s Park once extended southwards not only through the current Science Area, but some way south of South Parks Road, to the south edge of Mansfield College’s grounds. Along that edge, in the mid seventeenth century, a defensive fortification was constructed, to allow royalist Oxford to protect itself against parliamentary assault.

Holywell Manor, in its current incarnation

Mill stream and wells alike attest to the watery character of the land around the manor house.

It’s said there was a mill on the same site from the thirteenth to the late nineteenth century. It’s called Holywell mill in documents from the fourteenth century. It was rebuilt as a private house after successive occupants apparently struggled to make a going concern of it. The house’s residents have included the historian AJP Taylor and (briefly) as his lodger Dylan Thomas. Now it’s a Magdalen graduate house.   

Route along the north wall of Magdalen Grove towards old Holywell Mill site: Holywell Ford

The name of the mill implies that the holy well itself is older. As mentioned in the Groundwater posts (Uses and Meanings), it’s by no means clear which of several springs or wells historically found around here – at intervals between St Catherine’s and Holywell street – was the holy well.

Like so much of the rest of Oxford, this was clearly originally a pretty watery place, lying at a point where the gravel terrace on which the city rests slopes down towards the still shallower gravel base over clay along which the rivers run. Springs and wells presumably marked seepage from the gravel.

This watery resource has been exploited in various ways: not only to form a mill stream but also once a ditch around the town walls, then fishponds. Conduits were run from springs to supply St John’s Hospital  (predecessor of Magdalen) and Merton.

Initially this was an out-of-the-way area, outside the walls and not very close to any city gate, the nearest gates being a small gate at what’s now Catte st – at the western end of Holywell st, and the east gate, by the Eastgate hotel. Its relation to the city was ambiguous. Its church – whose oldest surviving portions date back to c.1200 —  was once a chapel of the city church of St Peter in the East. The church was appropriated by Merton at the end of the thirteenth century. But Merton leased out the manor, so there was an independent, historically mainly agricultural community. As the town crept beyond the walls, in the sixteenth century, and houses began to be built along the former town ditch, there were boundary disputes with the city.

Magdalen’s wall, which now dominates one side of the approach from the High Street, dates from the fifteenth century, from the early years of the college’s take-over of the Hospital of St John; so the wall antedates the facing houses. Once this route was flanked to the west by the city wall, and adjoining water.

City Wall bastion in New College grounds, visible behind houses in Holywell st

The bastion in Magdalen’s wall, facing Holywell street, marks the entry to Holywell parish. Holywell’s urban spur here shot off to the west – the heel on the boot, as I characterised it. This is clearly an urban street. It gained a touch of gentility, with the construction of Holywell Music Room in the eighteenth century. New College bought out cottages along the street to build a new front here in the 1880s.

But to the north, until the later nineteenth century, the land opened out. Behind the houses in Holywell street there was open land, bowling greens, gardens and meadows; then the Parks.

It was at the point where the urban met the rural that the space was most liminal: housing various appendages to the town, that were oriented to it, but for one reason or another, more easily placed beyond its walls.

Until the late eighteenth century, supposedly (on what evidence? evidence of maps?) a gallows stood by the bastion — in the middle of the road, one map suggests. (Oddly, it seems to me, since a gallows wasn’t usually a permanent construction, but something set up as needed). The gallows is said to have been Merton’s – they having claimed the right in the fourteenth century. But in 1589, it served as a place of execution for four Catholic martyrs, hanged for treason and felony (unlike Oxford’s long-celebrated Protestant martyrs, these have been commemorated by a plaque only in recent years).

More cheerfully, the area was a site of recreation. By the eighteenth century, there was a bowling green behind Holywell st, and another up St Cross Road on the other side (near modern Linacre?) There was also a solidly built cock-pit, where cocks were fought (and bets placed on their prowess).

In 1848, a site behind the St Cross graveyard was acquired to serve as one of three relief cemeteries (sister to cemeteries in Osney and Jericho).  

1852 saw the establishment in the manor of a Female penitentiary  and House of Mercy, run by an Anglican religious order, the sisters of St John the Baptist. In a list of such homes, it’s said to have offered 58 places to ‘Fallen women; [for] a year if possible; remain not less than two years’. It had no age limit. (Twenty years later, another temporary refuge for fallen women was established in St Aldates’, and at the end of the century, the Oxford Ladies Association established various refuges for friendless girls and single mothers). In 1929, this was relocated in Littlemore, and Balliol took over the house.

Until the 1990s, a site by the river was occupied by the Officer’s Training Corps.  When its long lease came up for renewal, the university seized the chance instead to use the site for a new Social Studies building.

Since the later nineteenth century, the main trend has been for this once open land to be colonised by the university and colleges. Merton, as the original landowner, has reserved a bit for itself – housing on Manor Road. But other land has been sold off to be developed for new colleges: off Mansfield Road, Mansfield College and Manchester College; off St Cross Road, Linacre College; off Manor Road, St Catherine’s College.

St Catherine’s College, water feature

The University Museum was the first outpost in what shortly became a much larger Science Area. Between Mansfield and St Cross Road, there’s a newish University Club building. Balliol and New College have exploited landholdings to construct new student housing: Balliol housing arises even now, in this period of rock-bottom interest rates, favourable to this form of investment.

South of the Parks themselves, such open land as survives mainly takes the form of playing fields – for Balliol, New College and Merton. Merton’s music and great meadows, described by Historic England as water meadows – both now in course of restoration to wildflower meadows — provide the chief remnant (if a re-imagined remnant) of an older landscape, along with trees along the river. For however long they may have been there.  Older photos show them, though the further back one goes, the harder it is to be sure….


Despite the sunlight and blue skies in these photos, as the bare branches suggest, almost all of these photos were taken during December and January 2020-21.

Groundwater: meanings

People manipulate the natural phenomenon, groundwater — a phase in the water-cycle — for various practical purposes. They also endow it with meanings. Meanings have changed over time, if through longer and more subtle processes than might have been supposed. In relation to springs and wells, themes of magic or miracle, health and good works recur, in changing guises. In relation to the logic of the phenomenon, providential themes endure, again in changing forms, through the nineteenth century. In later centuries, this theme co-existed with scientific forms of analysis and explanation. During recent decades, it’s repeatedly been said that understanding has just been revolutionised, but one might equally be struck by the long, slow evolution of evaluative tools and explanatory frameworks.

Holy wells

In his 1893 Legendary Lore of Holy Wells, Charles Hope listed and mapped ‘ancient and holy wells’, in Oxford among other places – stretching the ‘holy well’ category by blurring it with ‘ancient’. Of course, all kinds of wells can be credited with some loosely ‘holy’ characteristics. All springs can be seen as manifestations of God’s providence. Their appearance may be deemed miraculous. Insofar as it’s important that water sources should be pure, there’s reason to seek religious protection for them: the well is protected by its holiness. Varying mineral content gives a different character and taste to different waters, and some have gained a reputation as healing wells – an idea preserved in later ‘spas’ (deriving their generic name from Spa in present-day Belgium). The ‘wishing well’ idea perpetuates to this day the notion that wells and other water features possess quasi-magical properties. So there’s a cluster of possible associations, blurring the boundary between the holy and the workaday.

But some wells have stronger and clearer holy resonances than others.  Four Oxford wells stand out from the sources as sites of veneration.  St Edmund’s well; St Margaret’s well (also associated with St Frideswide); Stockwell by St Bartholomew’s hospital, and Holywell.

Sites of holy wells, plotted on 2nd series OS map, 1887-1914 (

Among these, two – St Edmund’s and St Margaret’s – were associated with specific religious narratives, and were in their day places of pilgrimage.

St Edmund’s well bubbled up in the parish of St Clements (probably now in the grounds of Magdalen College School, or possibly St Hilda’s). The VCH tells us that Edmund Rich, later archbishop of Canterbury, had a vision of the Christ child there when he was a student at Oxford. The spring began to be venerated shortly after his canonization (1246). (St Edmund Hall takes its name from the same saint, who’s supposed to have lived and taught on its site).

The well supposedly attracted hordes of pilgrims, such that one function of the hospital whose site Magdalen later acquired was to house such visitors (along with visitors to St Frideswide’s, later subsumed by Christ Church). Its water was reputed to cure wounds and sickness, but it attracted official church condemnation as superstitious at the end of the thirteenth century.

Nonetheless it continued to attract visitors until the early seventeenth century. Anthony a Wood (who offered an alternative account of its origins, linking it with the student days of Pope Innocent IV) thought that it lost its last shreds of popularity only when Milham bridge — which spanned the Cherwell between Christ Church meadows and St Clements, presenting foot passengers with an alternative to Magdalen bridge, and much used by ‘country people’ — was knocked down ‘by the violence of the ice’ in the great frost of 1634, such that the well ceased to stand beside a thoroughfare. This account, though, suggests that popular interest in it was by this point fairly casual.

St Margaret’s Well, a few miles up the Thames, at Binsey — still extant — was also an object of pilgrimage.

Legend had it that St Frideswide caused it to be opened up, when she prayed to St Margaret of Antioch to cure the blindness that had struck an unwelcome suitor pursuing her.  According to a Wood ‘To this Well and her Image, and Reliques in the Chapel, did the People come on Pilgrimage with as great Devotion to ease their burthened Souls, and obtain an Answer of their Doubts as they would to an Oracle. And here also, when the maimed or unsound had been cured by bathing in, or drinking of this Water, they hung up their Crutches as a special Memorandum of their Cure, for which Reason several Priests inhabited here, appointed by the Prior of St Frid[eswide] to confess and absolve them’. He claimed that pilgrim traffic from Eynsham across the hills to the west was heavy enough to support a flourishing hamlet at Sackworth, well furnished with inns, though subsequently decayed to rubble. A ‘little house of stone’ was set up around the well to preserve it from the throng. But this was taken down in the 1630s, possibly (the historian Alex Walsham has recently suggested) in the context of reviving religious tensions, which led some to fear that the achievements of the Reformation were in jeopardy.

Well of St Margaret at Binsey — Lewis Carroll’s ‘Treacle’ ie healing well

Well ceremonies associated with St Bartholomew’s chapel had a different trajectory – finding a new lease of life after the Reformation. St Bartholomew’s hospital (originally for lepers, later for the poor) stood on a little elevation athwart the watery slopes beneath Headington, near Cowley Marsh. According to a Wood’s Antiquities, it housed an image of St Bartholomew, besides which ‘many other Trinkets in the Chapel drew the Adoration of the People. In King Ed. IIId’s Time was here, St Edmund the Confessor’s Comb, St. Bartholomew’s Skin, the Bones of St. Stephen, and one of the Ribs of St Andrew the Apostle’. They ‘were on high Days indulged to view; and happy was he that could come near either to touch or kiss them. –Pilgrims came from afar to be cured by the Reliques,– Such as were troubled with continual Head-Achs, by combing their Heads with St. Edmund’s Comb, received Cure; such as had a Weakness of the Joints, by handling and applying these Bones to the Places affected, were restored to their pristine Strength; with many other like Accounts. …’  In order to attract visitors, and gifts for the lepers, Oriel – which had acquired the chapel and hospital — secured a grant of indulgences, but later in the century moved the relics to St Mary’s, to attract ‘a greater Conflux of People, than a retired and obscure Place.’  

‘Retired and obscure St Bartholomew’s, in Covid times

Still, the chapel remained a site of seasonal festivity: Cooks from Oxford flocked here, bringing in on Whitsun week the Fly. The Boys on May-Day the First-fruits of Flora, with their Lord and Lady’s Garland, Fifes, Flutes and Drums, to acknowledge with Dancing and Musick, and salute this gladsome Occasion.  And this Injunction and Custom was with great Earnestness and Zeal kept up by the Oxonians and the adjacent Country-Men, till the Reformation of Religion. When Q Elizabeth’s Act against Images &c. appeared, this Idol was pulled down.– Whence this Custom for a while slept, and the Alms-Folk by Degrees reduced to Poverty, and became the Objects of Compassion. …’

Yet, it seems not very long after, ‘the worthy Fellows of New College principally amongst others’, revived their former Ascension Day devotions’ (Ascension Day because Magdalen College Men and the Rabble of the Town came on May-Day). These involved fellows and their choir singing at the chapel, and (according to Hurst’s base text, Leonard Hutten) ‘haveing made their Oblations, and sung Anthems for a space, [concluding] this wholl Ceremonie and their Visitation with a passing along through the Grove to the Well, and doeing the like observance there.’ The well in question seems to have borne the name of Stockwell, like that by Hythe bridge (and suggesting mundane functions). The observance apparently came under some critical pressure, because whereas the choir originally ‘like the ancient Druids, echoed and warbled out from the shady Arbours harmonious Melody, consisting of several Parts, then most in Fashion.– [and on some later occasions] ‘sung an Oriana, or else one of Mr. J. Welby’s Songs of 5 Parts, beginning thus, “Hard by a Chrystal Fountain, &c.”, they subsequently reformed their ways and ‘sang only the Collect of the Day of  divers Parts; which done, they go up to the Grove’ – presumably by this means aiming to avoid any imputation of pagan or Popish well worship. In vain if so, because,‘the Presbyterian Times totally abolished it.’

In 2009 and 2013, New College briefly revived the custom. The then-choirmaster wrote an account of what transpired: ‘We … sang a short office in the exquisite chapel (with the East window bursting with morning light), and made our procession to the Well, at the top of the Oriel playing field. We were looking rather for a Spring, and in the wet season you could indeed find some water seeping out of the ground, if not bubbling. Little sign of it however on our bright if blowy Ascension Day morning. Unperturbed, we sang ‘Now is the month of maying’, a jollier number than Morley’s calculated ‘Hard by the crystal fountain’. The procession was led by pipe and fiddle, and we strewed the route with flowers as tradition demanded. Plenty of curious onlookers turned out to witness this spectacle: curious and genial’,

This was an assertion of the spirit of the place, intended to encourage respect for it (and to discourage encroachments).

In the case of Holywell, I have found neither reports of pilgrim hordes descending, nor of special ceremonies, though by report it did attract visitors. Indeed is said by Hearne to have brought ‘a vast amount of money to the place’ (what place?) No particular saint or story seems to have been associated with this well — or wells (for there was more than one in the Holywell neighbourhood).

Indeed, candidates jostle for the status of ‘the holy well’. One was incorporated under the altar of the manor house (perhaps when that was occupied, as Merton’s lessees, by the Catholic Napier family, champions of the old faith?). This well was rediscovered in 1897; a ‘tubular well stone’ found at that time was said to be Anglo Saxon work. Another adjoining on the exterior of the building apparently once served as a cold bath – perhaps this is the one reported to me to be still visible in the Praefectus’s garden. Another, called Jenny Newton’s well, lay or lies further to the east, and yet another beyond the bridge that now leads to St Catherine’s (called Napper or Napier’s bridge). Is St Catherine’s water feature fed by a holy well?    

According to a Wood or his continuator, whichever the holy well was, it was deemed ‘holy’ because reputed so; because its waters were often employed for holy uses, and because ‘certain Holy Men and Hermits’ chose to live in this ‘shady and arboreous place’. ‘About the year 1488, Dr Fitzjames, Warden of Merton, built a fair House over it of Stone, with a roof to it of Free-Stone, to receive the Prayers of People; a Token of which Bounty remains over the door of it at this Time, being a Dolphin neant, carved on a Shield, the Arms of this worthy Doctor.’ So not the one then under the altar, but one of the others.

During the seventeenth century, according to professor of chemistry and curator of the Ashmolean Robert Plot, it continued to be resorted to by people with bad eyes, which it was reputed to help. Again which of the wells exactly is in question is unclear.

Physic wells

Holy wells were often associated with miraculous cures. Belief in the curative properties of some springs or wells survived the Reformation, now being reframed in scientific terms. Indeed, interest in healing springs and associated ‘bathes’ or ‘spas’ took an upturn from the later sixteenth century, to peak (at least in England) in the eighteenth century. Having shrugged off one now-negative set of associations, springs became available for other kinds of appropriation. Their promoters often included local doctors who perhaps hoped to profit from attracting custom.

Springs were understood to acquire distinctive characteristics because of minerals in the rocks through which they passed. In the seventeenth century, the theme attracted the attention of chemists, who experimented to determine what was different about different waters, and speculated about how springs were generated and how precisely they acquired their different characters.  It’s perhaps not surprising that Robert Plot, the Ashmolean curator, who was not only generally curious about the properties of things but also specifically a professor of chemistry, interested himself in this debate (of which more in a moment) and, in his Natural History of Oxfordshire, drew attention to the reported healing qualities of Oxford’s holy well.

Crowell (outside the east wall of the old city, where Holywell st means Longwall st) is also reputed to have cured bad eyes.

The same was said of Child’s well — possibly adjoining Castle Mill Stream, though another Child’s well has given its name to the Chilswell valley on the slope above South Hinksey, and BBOWT believes that that was once thought to heal sick children and to have been honoured by Good Friday picnics – another tradition recently revived. It’s not always clear which traditions have been garbled or enlarged in the telling.

Of the Physic well further north along that range of hills, near Cumnor, south of Farmoor reservoir, Anthony a Wood wrote in June 1667: “This month about the middle, the well at Comnore in the high way going downe to Bablackhith was discovered and frequented. It will never be famous because there is not water to supply a multitude. Much resorted to by scholars; the water brought to Oxford.”   In 1806 it was said to be ‘long disused’ (Lysons, ‘Magna Britannia: Berkshire’). Henry Taunt photographed the well, noting of it “’In the Long Leys, on the way down to Bablock Hythe is what is called “Physic” Well which in 1674 [not clear why he picked on that year] was also much frequented by scholars of Oxford in search of a pick-me-up (it was reputed to have healing powers). It was here that the great cowslip grew that had three hundred heads’.

The route on which the well lies – a Wood’s ‘high way’ (and it is high) – survives as a bridleway link between two roads: Leys Road, Cumnor and a road at Bablock Hythe, whence a ferry used to cross the Thames.  According to Joan Tucker, the route remained long in favour for rambles from Oxford, so many must have passed by the well when in pursuit of other objects. It can still be walked – I met several couples and a family when I traversed it on a January Sunday afternoon — and the well survives.

Cumnor to Bablock Hythe walking route, from “nd series OS map (

A comparison between 2006 and 2011 photos on the Megalith site, and my 2021 photos of the Physic Welll underlines how much maintenance a site like this might need to be well-presented..

The ‘discovery’ theme in a Wood’s account is interesting. Discoveries had previously sometimes been framed as miracles. By the seventeenth century, not so, but could clearly still arouse interest. Plot was excited by news of a chalybeate spring (characterised by iron salts) being discovered near Osney bridge: ‘my worthy friend Dr Tho. Taylor has found so strong a Chalybeate Spring in Fulling-mill-ham-stream near Oseney Bridge [possibly the stream that now runs by Osney Mead], that notwithstanding last hard Winter (when the greatest Rivers were froze) this continued open and smoaking all the time, tinging all the Stones, by reason of its not running, nor mixing with other Water, with a deep rusty Colour’.

Explaining the water-cycle

What natural processes give rise to springs?

Two theories about the origins of springs were canvassed in the ancient world. One was that they came from rainfall — observation suggested some connection, since springs flowed more bounteously after heavy rain and dried up in droughts. Another theory was that they originated in the sea, seawater sinking into the earth and then percolating up through rocks, having changed character on its journey.

In the seventeenth century, these theories came back into debate. Plot referred to ‘this great controversy’. He was less interested in the position that some others continued to maintain, that intermitting springs portended dearths, wars, plagues, and other such prodigies. He professed himself ‘very diffident’ of this, and disinclined to discuss such notions further.

Hypotheses about the relationship between rainwater and springs rested on impressionistic judgements – more or less rainfall and more or less spring water. During the seventeenth century, work began on developing instruments to allow more precise assessments of these relationships: ‘Pioneers of the modern science of hydrology include Pierre Perrault, Edme Mariotte and Edmund Halley. By measuring rainfall, runoff, and drainage area, Perrault showed that rainfall was sufficient to account for the flow of the Seine. Mariotte combined velocity and river cross-section measurements to obtain a discharge, again in the Seine. Halley showed that the evaporation from the Mediterranean Sea was sufficient to account for the outflow of rivers flowing into the sea’.   

Stephen Switzer, whose summary of Halley’s calculations about the Mediterranean I’ve reproduced above, also nicely diagrammed the evaporation theory that Halley espoused, though he didn’t accept it himself, finding the rival ‘capillary’ theory more persuasive. Switzer is best known as a garden designer (he also served for a while as park quarries supervisor at Blenheim). His interest in springs arose from his interest in fountains and other decorative ‘waterworks’ (all the rage in the formal gardens of the seventeenth and early eighteenth centuries – as in Oxfordshire notably at Hanwell).  Switzer’s case nicely illustrates the diversity of preoccupations and backgrounds from which people might be drawn into this field of enquiry.

By the end of the eighteenth century, the evaporation theory seems to have carried the day, though I’m not sure how that happened.

Hydrology and hydrogeology

During the eighteenth century ‘hydrology’ caught on as a name for the science of water. It was established as a branch of natural history  and as a topic in applied mathematics.    

Thence to the end of the nineteenth century, several lines of development interacted.

It had long been clear that understanding groundwater depended on understanding the earth, or geology, and what would come to be called ‘hydrogeological’ processes. Late eighteenth and nineteenth-century developments in geology accordingly provided important context for new approaches to groundwater. William Smith, maker of the first geological map, explored the relation of spring lines to stratigraphy, and drew on his conclusions to advise landowners and canal companies. Conybeare and Phillips’ pioneering early nineteenth-century geology text book tried to make sense of springs and wells, and developed a technical vocabulary, distinguishing ‘porous’ from ‘impervious’ strata. In his Bridgewater Treatise, Geology and Mineralogy considered with reference to Natural Theology, (1836), the Oxford geologist William Buckland provided a notably clear explanation of the principle of the artesian well, until then known only to specialists: that is, a well which taps trapped groundwater, which rises within the well in response to pressure in the underground trap.

Alongside geological advances, techniques for measuring water volume and flow continued to develop, and a bank of data about British rainfall, springs and rivers was built up.

John Dalton remains a name to conjure with. His work built on a tradition of quantitative work on weather, and was written up in his Meteorological Observations and Essays.  In the words of Nathaniel Beardmore’s early manual of hydrology, Dalton ‘investigated the subject of the relative fall of rain and supply to springs and rivers…..he contrived the percolating gauge, and made extensive experiments on the subject from 1796 to 1798 at Manchester’. He is celebrated now as having been ‘the first to express the basic components of the evaporation process from a free water surface in quantitative terms.’ But Dalton’s law wasn’t expressed by its author in mathematical terms; that came later. Although quantification was in vogue and advancing through this period, mathematicization was only incipient.

Dalton’s gauge, fro the Illustrated London News 1876, and tthe title page to an 1834 reprint of his essays

Beardmore’s mid-nineteenth-century manual started out as a book of tables: both tables useful in calculation and tables of quantitative data. Including for instance rainfall data for different parts of the country:

From Beardmore’s Manual of Hydrology (

This table covers a year of very heavy flooding: 1852.

Later commentators had a lot of quantitative data at their fingertips.  When the geologist Joseph Prestwich (in his Letter on the Oxford Water Supply, 1874), considered the option of redeploying the spring at Hinksey (which had once fed the Carfax conduit, and the city’s early modern piped water system), he statd it yielded 10,000 gallons a day, and could easily be made to yield more. He said that he regretted that even the nearest Cotswold springs, near Fairford, were too distant to be usable, for they yielded no less than 4-6 million gallons a day. In the following decade, de Rance, who worked for the British Geological Survey, in his book on The Water Supply of England and Wales, cited a civil engineer on ‘the summer delivery of the Isis at Oxford’ being 73m gallons a day, and in winter 320m gallons.

Hyetographical’ (rainfall) map from de Rance Water Supply

Hydrogeological mapping was an offshoot of geological mapping – but such maps might also include other kinds of water-related data, including quantitative data. Prestwich has been hailed for having, in 1851, produced ‘the first geological map that included groundwater information’ (in relation to possible sources for the supply of London). Charles Lucas, a staff member on the British Geological Survey, not only coined the term ‘hydrogeology’ but also ‘produced the first real hydrogeological map’ in 1877.

Advances in chemistry made a contribution: ‘Analysts became more confident in their determination of the constituents of water.’

Religious perspectives were not initially superseded by these scientific developments.  The tradition of natural theology, which took the hand of God the designer to be manifest in the natural world, informed the presentation and perhaps helped to motivate some of these enquiries. Buckland’s Geology and Mineralogy was one of a series of Bridgewater Treatises, focussing ‘On the Power, Wisdom, and Goodness of God, as manifested in the Creation’. Nor was this perspective confined to men trained in Church-linked universities.  The civil engineer Beardmore, in his manual, wrote of wells that they were ‘provisions for the wants of humanity, ordained by providence, so that man shall have water in detail where he may require it for his daily use…the chemist informs us that carbonic acid and other minerals in the natural soil will constantly purify the percolating water in the slow passage (filtration) downwards…so that the earth, synonymous with corruption, is also all-powerful in production and support of new life, as described by the great apostle before chemistry was known.’ The role of modern applied science was, in his account, to compensate for civilisation’s unbalancing of this natural order: ‘Wherever immense populations are gathered together, these conditions are interfered with so as to upset the ordinary economy of nature, and give rise to the complications which the engineer is called on to adjust.’

Developments in scientific understanding made it possible to push God further and further into the background in explanation; they didn’t of themselves require that he be expelled.

Developments in knowledge and understanding were deployed in practice in a series of mid and later nineteenth-century discussions and debates around public policy, above all around the question of how to keep increasingly densely populated towns supplied with sufficiently pure water. There was more to that debate than just an understanding of the production, location and destination of groundwater, but changing ideas about that informed policy debate, and fresh work was also stimulated by it.

Geologists with Oxford connections of one kind or another were among contributors to these debates. These — like the Oxford geologists already cited — are all men we’ve met before (in my posts on Geomorphology, Dinosaurs or Dredging). 

The Rev James Clutterbuck, inspired as a student by Buckland, interested himself especially in water supply and came to prominence for his role in a fierce argument about how best to supply London. A leading historian of British hydrogeology says that one of his achievements was to recognise that groundwater was a finite resource – though this seems implicit in older discussions of its origins in rainwater. ‘He was the first British worker to apply observations of groundwater levels in a practical and innovative way to the study of groundwater flow.’ So perhaps it was his ability to give practical form to his ideas that was crucial.

Joseph Prestwich, initially like Clutterbuck a keen amateur (he made his living as a wine merchant) also interested himself in the challenge of supplying London, and wrote a major book on the topic, his Geological Enquiry into the Water-bearing Strata of the Country around London (1851). He was made a member of the Royal Commisson on the Pollution of Rivers in the 1860s, serving alongside (among others) the engineer Nathaniel Beardmore. He became Oxford’s first professor of geology in 1874 – the year in which he published his pamphlet on Oxford’s water supply.

None the less, there remained significant disagreements, especially between geologists and engineers, about how much underground water was available. Public policy debate brought people with various forms of expertise to a common table; it didn’t ensure that they would resolve their differences.

Hydrogeology now

Twentieth and twenty-first-century hydrology has continued to build upon and expand these inherited understandings, techniques and databanks (though is much less likely explicitly to invoke providential design).

There were centuries of cross-fertilisation between British and European work in the field. Latterly, concepts and techniques developed in America also began to shape British research and practice (‘aquifer’ was an American coinage).

Look backward from any point, and the advance in knowledge and understanding always seems notable. Look forward, and what strikes commentators is what, at any point, had yet to be achieved. Thus, it’s noted that, in Britain, ‘[T]he collection of rainfall data was not taken over by a Government Department until 1919, and it was not until 1935 that the systematic collection of data on both surface and groundwaters finally began’.

Since 1945, developing national legislative and governance frameworks have shaped British research and practice.  The BGS map portal gives access to the latest version of the official British hydro-map.

Meanwhile, computer assisted modelling has helped modern hydrogeologists to cope with the extreme complexity, variability and even randomness which are now stressed to be features of water systems.

Thanks to Benjamin Thompson and Nancy-Jane Rucker for information about New College’s recent revival of the Ascension Day ceremony at Bartlemas and its well, in which both of them played a part. Also to Stephanie Uenkins for finding Anthony a Wood’s account of the ‘discovery’ of the Cumnor Physic Well, which is in his Diaries.

Alexandra Walsham has interesting things to say about holy wells and physic wells, in her Reformation of the Landscape. In relation to physic wells, I have also benefited from reading Noel G. Coley, ‘”Cures without care”: “Chymical physicians and mineral waters in seventeenth-century English medicine’, Medical History, 1979.

.I was put on to Switzer’s illustration of Halley’s theory by Asit K. Biswas, ‘Edmond Halley, FRS, hydrologist extraordinary’, Notes and Records of the Royal Society of London, 1970.

For the history of hydrogeology, I have found especially useful works by John Mather (several of which are also available online to Bodleian readers): see eg ‘From William  Smith to William Whitaker: the development of British hydrogeology in the nineteenth century’, in DJ Blundell and AC Scott eds, Lyell: the Past is the Key to the Present, 1998; ‘National water supply, debates between geologists and engineers and the role of the Society of Arts’, Earth Sciences History 2018 (I don’t have access to this, but a summary can be found here); and J.D. Mather ed., 200 Years of British Hydrogeology 2004.

Joseph Prestwich’s works cited here are available online to Bodleian Library readers.

Gulls in recess

(The Madwoman turns, watching the flight of the birds.) Ferryman, there the wild birds float! I see the wild birds fly! What are those birds?

Ferryman Those? They’re only common gulls.

Benjamin Britten, Curlew river   (text)

In the Japanese Noh drama from which Britten took his plot line, the birds are gulls. His version locates them on the ‘Curlew river’. But curlews don’t fly and float around the water: they’re wading birds. The birds the madwoman sees are surely gulls, as the ferryman tells her. But for her, gulls it seems are not resonant enough. She wants them to be curlews. Humouring her, her auditors are gradually drawn into her world.

There are always gulls around the Thames. Commonplace birds: one doesn’t register them much, though in the summer I took some footage of them swooping around – wild birds flying — presumably catching insects over the water.

It’s in the winter that whole gull flocks appear around Oxford rivers, as they leave their summer haunts along the coast to take refuge in the warmer interior. I first started noting their growing numbers in October.

October. First as single spies, and then in battalions

A friend in Exeter told me that, at the same time, their numbers had noticeably shrunk in his neighbourhood.


It seems some gulls change their colouring in winter. Here are some black-headed gulls in their winter dress: as if with snow hoods on.

The arrival of gull flocks puts gull behaviour more noticeably on display. Mainly one sees them floating or flying.

Their habits are their own: contrasting with the habits of the more enduring swans and geese. Just as on a jungle safari you can watch different animals living out their different lifestyles (they do this in real life, not just on TV programmes) so around the Thames in winter you can contrast a whole series of avine sociabilities

Geese spend at least as much time on land. Not so gulls. They seem to spend as much time on the water as they can. Though they also like perches just above the water. (Geese like to eat grass; gulls are omnivorous, and also like insects, fish and small marine life, which they may stir up from the water with their feet).

Contrasting behaviour of geese and gulls

Gulls foraging and flying

And whereas geese have accustomed territories (even though not all local residents; some are winter immigrants), gulls seem contrastingly opportunistic, rapidly colonising any new stretch of water that rain and flooding bring into being.

Gulls colonise flooded land, jointly or singly

Though gulls often rest immobile on their perches, or float around placidly on the water, they are also easily roused to flight, pursuing rumours of bread, then falling back disappointed; or swarming above some incident on the land.

Watch geese and you realise there’s some goose soap-opera going on, with lots of short-term drama, though the plot may be opaque. Gulls don’t seem to row with one another as much. Floating, they don’t seem to socialise: they’re lonely in crowds. But they excite in groups. They’re not prima donnas but the chorus, responding, commenting.

You’ve been there ages. It’s my turn

Cowley marsh

Old Cowley marsh was a wedge-shaped area, tilted roughly NNW to SSE, straddling what’s now the Cowley Road, extending to Bartlemas chapel on its NE edge, what’s now Cricket Road to the WSW, and the Boundary Brook to the SE. Or that’s how it’s shown on a 1777 Christ Church estate map.

Cowley parish, based on Christ Church 1777 estate map

Now adjacent but variously described portions of this land survive as open land for public use. Cowley Marsh recreation ground is the largest fragment. Other odd parcels survive at the NE and SE edges of the recreation ground, around Barracks Lane. Barracks Lane Meadow lies to the north east. To the south east until recently lay, south of Barracks Lane, a wildflower meadow, now much truncated, and to the north – perhaps historically ‘Lye Hill’ rather than Cowley marsh — a triangular Cowley Marsh Nature Reserve.

Truncated remains of wildflower meadow, new waste depot space beyond

Together, these properties occupy a significant chunk of what was once the SE quadrant of the wedge. Much of the wedge northwards on both sides of the Cowley Road has been covered with housing, while Motofix and Elder Stubbs allotments occupy the SW quadrant.

Though a NW part of the marsh was in use as a Magdalen cricket ground by the early nineteenth century, most erosion of the nominal ‘marsh’ area probably followed Cowley’s enclosure in the 1850s. The second series OS map already shows the two northern and the south-western quadrants marked out into plots, though at that date (late nineteenth/early twentieth century) the plots still lay almost entirely open. On this post-enclosure map, what survived of ‘Cowley Marsh’ was a ‘recreation ground’ (not on its current site, but to the south of the Cowley Road), and, on the site of the modern recreation ground, a ‘cricket ground’.

Cowley Marsh in transformation: 2nd series OS map 1888-1914 (

On all but the east side of the pre-enclosure marsh, as shown on the 1777 map, lay the strip fields of Cowley parish (it’s not reported what lay to the east, because that was St Bartholomew’s extra parochial area, to which the map doesn’t extend). There’s some indication in place names that westwards – below the marsh — there were further soggy areas. The fields to the west are called ‘The Lakes’. If this is to be understood in the old sense of ‘ditches’ (there are no evident lakes in the modern sense), then it may have been drainage ditches that kept this land dry and cultivatable. Further along the course of the Boundary Brook, as it ran westwards from the marsh towards the river, we find Sanders Marsh. (I first thought that ‘Meers’ in the same area were pools, but in fact they were embanked boundaries. There were more than a dozen other meers scattered across the parish. Boatmen apparently knew underwater ice ridges as ‘ice meers’.)

It’s not hard to see from the topography why Cowley Marsh was marshy ground, though I can’t myself distinguish between ways in which the landscape shaped the flow of water and ways in which the flow of water shaped the landscape. Still, the main outlines seem clear. To the east of the marsh, the land slopes up towards Headington. Lying at the bottom of the slope, the marsh occupies a depression (the result of past, more voluminous water action?) with a base of ‘mudstone’. The lower slope above is sandstone. To the SE, there’s another small sandstone slope that raises Church Cowley and Temple Cowley to a greater elevation. Geologists call this sandstone ‘Temple Cowley member’.  On the higher slopes lies ‘Beckley sand’.

To the NW, as one follows converging roads towards St Clements, gravel terraces also slightly raise the level of the land– on the 1777 map, ‘Ridge Field’ lies NW of The Lakes (around modern Ridgefield Road). Above the sandstone on the slopes up towards Headington lies limestone, capped, as you keep ascending, by other rocks.

So water falling on the hills would soak through the limestone and sandstone and then issue out. The Boundary Brook and Lye valleys take their rise in Headington, run their course down the sandstone slopes, fusing with one another above the marsh and then running on towards it.

Paralleled, no doubt, by more diffuse seepage from across the limestone/sandstone front. Robert Plot, the first Ashmolean curator (in its early, science-museum days) characterised such seeping waters thus: ‘of so slow a Pace, that they seem rather to sweat than run out of the Earth…. [they] are stopt upon the very Surface.’

This 1868 British Geological Survey map does not precisely anticipate modern geological mapping, but it does suggest something about the situation of the marsh.

Geology of the area around Cowley Marsh,
Contains British Geological Survey materials © UKRI

Bartlemas chapel and hospital, at the NW end of the marsh, and not historically included in it (they formed a little extra-parochial area of their own) lie broadly within the same zone. There was once a spring or well around there, probably at the point where the sandstone gives way to Oxford clay, in what used to be the northern part of the Bartlemas extra-parochial enclosure, now some way into Oriel’s playing field. Meanwhile, the allotments which now spread out beneath the chapel are moistened (and sometimes more) by water seeping from the ground.

In principle, another spring line should lie along the edge where the sandstone which underpins old Cowley encounters Oxford clay. Crowell Road seems not to extend sufficiently far down this slope to hit this front – but it may of course once have headed in the direction of a well frequented by crows. If the ‘Catwell’ rose where Catwell Close now stands — it serves new housing that’s recently intruded on to the west side of the recreation ground — then that may mark a point at which water seeped out of a gravel bank.

The marsh’s physical characteristics have long given it a distinctive role.

As I’ve already noted, Cowley common fields were historically distributed around three sides of the marsh, bordering on, and to some extent mixing in with, those of St Clements to the NW. The marsh provided common pasture land – as riverside land (most notably Port Meadow) sometimes also did. On a 1797 map, the marsh is termed ‘Cowley common’.  St Clement’s parishioners shared rights in a chunk of this marsh-pasture.

Cowley’s initial areas of settlement – Church Cowley around the parish church of St James, and Temple Cowley — lay on the sandstone to the SE. Church Cowley is an ancient settlement, listed in Domesday. Temple Cowley developed around the Templars‘ preceptory in the twelfth century.

St James, Cowley; barn on site of Templars’ preceptory and eighteenth-century house, both on Temple Road

Routes to London traditionally passed either side of the marsh, flanking the side of Rose Hill or climbing through what’s now South Park to proceed along Old Road through Shotover.The marsh could be skirted, wandered across, or crossed via a causeway that presumably provided the most reliable route, being elevated above the sometimes wet land.

Of the two historic Cowleys, Temple Cowley most nearly adjoined the marsh, which lay between it and the city. In the eighteenth century, there was a peat-digging industry focussing on the marsh, run from Temple Cowley.

In the early nineteenth century, William Turner of Oxford painted the view towards Oxford from Bullingdon Green (now the eastern end of the Oxford Golf Course), looking over an open and watery Cowley marsh.

Following enclosure in the mid nineteenth century, as building moved outwards from St Clements, a new NW Cowley urban area emerged, around the St Clement’s end of the Cowley Road, which was in 1868 organised into a distinct parish, Cowley St John. Cowley Marsh was allocated to this new parish. But it and the area to the west of it were relatively slow to be built upon, wedged between the old settlements and new suburbs, and moreover marshy.

Reclamation of the marsh started in the middle ages. The VCH suggests that there was drainage in the thirteenth century, probably organised by the abbey of St Frideswide’s, releasing some formerly soggy land for arable — would that be the Lakes Field? A new round of drainage probably accompanied mid-nineteenth-century enclosure. The Boundary Brook – which is now a complex stream, sending out offshoots in various directions, seems to have provided the main drainage artery. One branch runs along the top of Cowley Marsh recreation ground, presumably channelling seepage from the hillside above.

There’s a long history of sport and competitive games in and around the marsh – up the hillside to the SE, that is on Bullingdon Green (now part of the Oxford Golf Club) and on the marsh itself. Colleges, University, the City and local people divided and shared the space. Already by the early nineteenth century Magdalen was using land on the NW edge of the marsh as a cricket ground. That also served 1851-81 as the University Cricket Ground (before that was relocated to the Parks). Old marsh land to the SE (in and around the current recreation ground) was used for cricket and golf by the 1870s: various colleges had cricket pavilions on the ground, and the University Golf Club also played there, as did the Cowley St John neighbourhood cricket club. The Cricketers Arms in Junction Road (which continues Temple Road towards Barracks Lane) and Cricket Road (along the south edge of the old marsh) commemorate the game and its players. In the 1890s the City appointed land adjoining Magdalen Cricket Ground as its first ‘public recreation ground’ (as authorised by legislation from the 1870s, on which they’d been slow to act). In the 1920s, the University purchased Southfield Farm, on the hillside above the marsh, and this provided the basis for a more extensive golf course.

Erosion of the green space has continued, as pressure for housing makes land that once looked suitable for utilities look too good to waste. A bus depot that was placed along the Cowley Road in the 1920s has been replaced (since 2004) by housing. The evidence of maps suggests that the waste depot that now occupies the SE end of the old marsh wedge was first established there between the wars.

View towards the waste depot

Intensified interest in preserving a green environment has seen both wins and losses. The Cowley Marsh Nature Reserve occupies a water-bounded triangle extending from the SE corner of the old marsh; on post-war aerial photos, this seems bare. According to the Council, it now ‘combines wet grassland with small patches of wildflower meadow’. It has increasingly been treed over. Henry Taunt’s photograph of the boundary stone on the marsh in 1914 places it on relatively open land, backed at a few yards’ distance by a line of large trees. Now the setting is completely different. Along Barracks Lane run a row of mainly small trees and bushes, while behind stand a series of coppices: hazel and birch.

In 2009, the Council sowed wildflowers alongside a children’s playground at the Barracks Lane side of the depot. But in 2015, it moved to extend the depot site over this space, and prevailed against vigorous local opposition.

If one consider the whole swathe of land that runs up and down the hill from the marsh — passing through the depressed land on the LIDAR map, including space once occupied by Lakes Field on the way to the river — it’s striking how much of it remains open, if not always open to the public. On the hillside above the marsh lie the golf course and the grounds of Oxford Spires Academy (preceded on this site, from the 1930s, by Southfield Grammar School). Then comes the recreation ground, then Motofix and Elder Stubbs allotments (established here by way of compensation when Elder Stubbs up above the woods was enclosed). Then St Gregory the Great school and East Ward allotments (and on the other, SE side of the Boundary Brook, Florence Park). Then the Iffley Academy; then various boating sites, meadows and nature reserves. All interspersed with patches of housing. This pattern no doubt testifies in part to the late development of this space between old Iffley and Cowley and the expanding city, but also to the wateriness of the ground, suggesting the wisdom of leaving much of it open.

I’m not sure how much it reflects an older pattern, and to what extent it points to the success of mid nineteenth-century drainage efforts, but current flood-prone areas lie not squarely on the line of the marsh, but rather on both sides of the Boundary Brook.

Marsh Harrier pub in Marsh Road. The marsh harrier still appears on county bird lists. But it seems doubtful that the marsh as it is now is its natural habitat.

With thanks to Robert Quick for the loan of his camera when the cold weather froze my iPhone.

Groundwater: uses

Rain falls on the hills and the plains, and drains into rivers and brooks. Or else it pools within permeable rocks, forming a ‘water bed’, or ‘aquifer’. The surface of the saturated area is the water table, which, the National Geographic tells us, ‘unlike the tables you’d find in your house… usually isn’t flat, or horizontal’, This stored water sooner or later pours or seeps out of the ground, in the form of more or less ebullient springs, at places where gravel, or chalk or permeable rocks meet clays or less permeable matter, that is, at the ‘spring line’.

Or, as the first Ashmolean curator Robert Plot expressed it, in his Baroque prose, springs take the water, ‘that we see they dayly vent, from Rains, Mists, Dews, Snows, Haile &c. received into the Spungy tops of Mountains and sent forth again at the feet of them, or somwhere in their declivities’.

We now usually derive the water we use for washing, cleaning or drinking from captured, piped water. But in the past, springs or wells (which tapped subsurface water) played a more obvious role in life. Plot (1686) thought of springs as something that ‘we’ see vent ‘dayly’, and, almost two hundred years later, in his Manual of Hydrology (3rd edn., 1862), Nathaniel Beardmore took it to be common knowledge ‘that ordinary wells fail in summertime, and … few wells have a never-failing character’.

Springs and wells figure in numerous of Henry Taunt’s photos of the countryside around Oxford (though this may reflect his particular interests) . A Historic England image search on ‘Taunt well’ and ‘Taunt spring’ produces lots of regional images, including numerous of St Edwards’ well, near Stow-on-the-Wold; but also of wells in Hook Norton, Chipping Camden, Guiting Power, and Aldworth, West Berkshire.

English hill-dwellers sometimes lacked easy access to water. Much depended on local geology. In the 1860s Rev James Clutterbuck, a passionate student of geology, who had studied with Buckland at Oxford, and had a special interest in water supply, noted that on chalk hills, impermeable layers sometimes trapped water close to the surface, so that it could easily be recovered, encouraging hill-top settlement. Nonetheless, in such districts ‘a rigid economy in the use of water is forced on the agricultural and other populations, as, on the failing of the supply, no resource is left but the streams in the valleys, or very deep wells’ He had in mind places such as Ipsden in the Chilterns, where, according to the VCH, ‘water supply until the twentieth century was from [shallow] wells, ponds, and rain-water tanks, resulting in shortages in dry conditions: at Stoke Row [it was] fetched from ‘dirty ponds and deserted claypits’, and passed in times of drought from cottage to cottage for cooking’’  Clutterbuck (and the VCH) note the solution ultimately found: ‘an Indian magnate, the Maha Rajah of Benares intrusted to my friend, Mr EA Reade, C.B., a sum of money to be employed for the benefit of the poor, on an object not of a religious character. He expended it on a well in the chalk, 358 feet deep, furnished with simple but most serviceable machinery for raising the water – large buckets, chains passing over iron sheaves, wound up by a winch and flywheel’. The VCH notes that  ‘Another Indian ruler friendly with the Reades paid for a 125-ft well at Ipsden church in 1865.’

Oxford doesn’t suffer these chalk hilltop problems. It receives its waters from far and near.

Springs and rivers

The Thames and Cherwell originate in springs: the Thames at Thames Head, in Gloucestershire, and in the numerous other springs elsewhere, which poured into its tributaries. There springs have their source in aquifers in the Cotswolds – from the great belt of oolitic limestone that runs NE/SW, and from the cornbrash intermingled with it. Some of these spring waters then run as rivers through the clay.

The photographer Henry Taunt – such a fan of the river that he renamed his house Rivera —  certainly seems to have made a point of visiting and photographing this set of springs, often repeatedly. Eg at Thames Head; at the source of the River Leach; the Windrush, and the Churn. His enthusiasm didn’t extend to drinking unfiltered Thames water, nonetheless: he pressed for its improvement, demonstrating that what came out of the Oxford water company’s taps contained shrimps as well as plant matter.

The Thames remains the main source of local drinking water, now purified naturally by passing wetlands at Pinkhill Meadows, on the western arm of the Thames, thence being pumped up and run through filters into Farmoor reservoir, where its quality is monitored. It’s then further filtered and treated before being passed on to customers.  A report on a field trip made by the Oxford School of Geography and the Environment provides a good account of how Farmoor works.

Farmoor Reservoir and control tower

Canal reservoirs

Where does canal water come from? (a good question that I hadn’t thought to ask until it was put to me). The Canal and River Trust states that ‘When the canals were first built over 200 years ago, finding a reliable water supply to ensure that freight could be transported around the country was critical to the success of those early canal companies’…Canals don’t discharge into the sea, but they lose water to evaporation and seepage. And when locks are opened, waters move down canal. ‘Supplies of water come from a network of reservoirs, rivers and streams, as well as being pumped from underground.’  Water supply has to be calibrated to changing need.

The Oxford canal more particularly gets topped up from three reservoirs above Banbury, around the canal summit (so that gravity can carry the water down): Clattercote and Wormleighton (both established in 1787) and Boddington [completed 1805].  Clattercote was once the site of a priory, whose lands were granted to Christ Church at the Dissolution. The priory’s leper pond became a fishpond, which formed the basis for the later reservoir.

James Clutterbuck noted the favourable geology of the district where these reservoirs sit: ‘the base of the valley is the lower lias [clay;]…the intervening marlstone, usually charged with water, adds another source of water’.

The engineer Nathaniel Beardmore told the 1866 Thames enquiry that he thought that if the Thames around Oxford were to be improved for navigation, reservoirs might need to be constructed to keep it topped up as needed, in similar fashion. He said that in his view ‘the portion of the valley above Oxford is destined some day to have large reservoirs’.

Springs, brooks and fens

Around Oxford, the rivers receive more run-off from brooks and groundwater originating in another aquifer NE/SW, this one composed of looser Corallian rag, rich in fossils. Local springs arise on the surrounding hills at points where permeable rock rests on Oxford clay, though rocks may be layered and springs follow accordingly complex patterns.

Robert Plot, in the late seventeenth century, noted springs on the Headington side of Shotover, which rose out of the ground and then quickly disappeared back into it. An image of a ‘pothole spring’ on Headington Moor is preserved in one of Henry Taunt’s photographs.The British Geological Survey’s ‘Geology Viewer‘ suggests that around here, limestone gives way to clay. Here one finds ‘Spring Lane’ (but the Shotover reservoirs, on the Kimmeridge clay that tops the limestone, have water pumped into them from elsewhere).

Somewhere to the south of here lie the origins of the Lye Valley (though a recent geological surveyor failed to find the Lye’s exact source).

Of the springs on the hills around Oxford, some give rise to superficial patches of wet ground: thus on the hillside above Botley, now the Fogwell estate.

Cumnor slope ravine, from Google Earth, data from Landsat/Copernicus, spring 1984

Other springs give rise to brooks which sometimes carve deep channels into the relatively soft stone as they descend: thus especially the ‘ravines’ on the hillside beneath Boar’s Hill, but also the Lye valley. These channels may be associated with fens. ‘A fen is a type of mire which receives at least part of its water and nutrients from soil, rock or groundwater… Fens contrast with bogs which mainly receive water and nutrients from rainfall alone’.

Many Oxford brooks pick up lime from the base stone, and accordingly have the potential to nourish alkaline fens. This is the case with the various watercourses associated with the Wild Oxford project (Chilswell, Raleigh Park, the Lye Valley and Rivermead) as well as the Hinksey Heights Nature Trail all of which Judith Webb has played a part in restoring – though she expresses more excitement about the large rich areas of fen around tributaries of the Ock around Sunningwell and Sandford.

Marley Fen in Wytham woods is another alkaline fen.   

Alkaline fens may form ‘tufa’ deposits. Robert Plot noted this phenomenon. ‘In the Parish of St. Clements in the Suburbs of Oxford, about a quarter of a Mile distant on the right Hand of the first Way that turns East-ward out of Marston-Lane, there is a Ditch, the Water whereof incrustates the Sticks that fall out of the Hedge’ (though he thought this ‘inconsiderable’, scarcely worth mentioning).

Having the potential to produce alkaline fens doesn’t mean that fens necessarily result, or survive. Urban development, changing the character of ground cover and drainage, or introducing new elements into the water flow, may affect them, as may tree and plant growth that harvests the water. Preserving these fens involves preserving the water flow, maintaining its chemical composition, and stripping away some trees and other vegetation.

By contrast to these alkaline areas, Shotover supports acid marsh and grassland.

Wellhouse, Wick Farm -spring fed from hills above Barton

See also the ornamental well-head at Shotover Park, presumably fed from the heights of Shotover.

City wells, springs and pumps


Down in the valleys, the clay over which the rivers run keeps rainwater close to the surface, where it pools in gravel terraces. Clutterbuck suggested that the ready availability of subsurface water, which could easily be tapped for wells, will have been one reason why the site of Oxford (like the site of London) attracted early settlement.

Modern archaeologists surveying builders’ trenches commonly find well pits associated with late Saxon house sites – thus in (what’s now) Queen st and Cornmarket. Later wells (for example to the northeast of the north gate) provide an indicator of the expansion of settlement. It seems that wells aren’t always easy to distinguish from simple pits, though one assumes the wells will have been a bit deeper – or else the pits wouldn’t have been good for much. Water-filled pits could perform other functions. There’s a nineteenth-century instance of a backyard fishpond in St Ebbes.

Such traces have long been recognised for what they are. In the seventeenth and eighteenth centuries, people digging for gravel ran across ‘Tokens of Trenches, and round hollow places in the fashion of wells’ and knew that they were looking at ‘Vestigia of venerable Antiquity’ – in this case, of Beaumont Palace, which, between the twelfth and sixteenth centuries, when it was finally demolished, stood around what’s now St John’s Street.

A few institutional wells survive: a rather grand ‘well chamber’ was built into the motte (the mound) at Oxford Castle in the thirteenth century, the motte being raised higher to house it. Perhaps implying that the ten-sided stone keep, which survived until the eighteenth century and was presumably served by this well, was only built after this time. The chamber lies 20 feet below the surface. Henry Taunt’s collection includes a photo of carved panels in this clearly quite grand well chamber.

Not an obvious site for a well

St Edmund Hall proudly displays a thirteenth-century well, rediscovered only in the early twentieth century, Merton has a flagstone marked ‘well’ in the northeast corner of the front quad.

For how long did wells continue in use, and what proportion of houses had their own wells? As the surface of the city rose over time, ground water presumably became less and less readily accessible (though still probably only a few metres down). A Christ Church 1829 list of tenements (published in the Oseney Cartulary) doesn’t list wells among appurtenances, though it does list privies and washhouses, and sometimes pigsties. Were wells not present, or did they just not merit inclusion in the list? Does ‘wash house’ – a laundry shed – imply a well? We learn from other sources that the poorest lodgings didn’t have access to water. When a scheme to provide a public wash-house and baths for the ‘industrious classes’ of Oxford was launched in 1850-, its supporters said that 600 families in the western part of the city had no provision whatever for bathing or laundry. An appeal for funds to establish public facilities bore fruit in 1853, but the enterprise did not pay its way, though it limped on until 1867. Bigger houses, with more ground of their own, were probably more likely to have wells. They remained common enough in the 1860s for Clutterbuck to relate that a new drainage trench cut through corrugations in the clay at Oxford had laid dry ‘the house wells…through a considerable district’.

(On the Berinsfield website – a new village SSE from Oxford — there’s an account of gravel diggings in the post-war period draining water from local wells and making it necessary to deepen them).

Extra-mural springs and wells

The spring/well distinction is by no means clear-cut in the case of shallow wells. Indeed, the Old English ‘welle’ meant a spring.

Medieval proto-surnames, as they appear in the Oseney Cartulary, quite often made reference to what were presumably neighbourhood springs or wells, doubling as landmarks, thus ‘de Bradewell’, ‘de Bugwelle’ ‘de Cokeswell’ ‘de F(o)ulewelle’ ‘de Fretewell’ ‘de Haliwelle’  de Hertewell’ ‘de Lodewell’, ‘de Siwelle’ ‘de Stokwelle’ ‘ ‘de Warmwelle’. Also Attewelle.  The majority of these don’t relate to later known wells, though ‘Haliwelle’ and ‘Stokwelle’ are exceptions.

The public wells which attracted notice in medieval and early modern texts, and which appear in modern lists, seem essentially to have been springs, around the margins of the city, or further out. Their framing for use could be rudimentary: Walton Well, as shown by Henry Taunt, was a hole in the ground framed by an oblong of stones (this was shortly before it was channelled into a slickly presented ‘fountain’).

They lie in areas close to where we know watercourses ran, and it’s easy to imagine that water might have been quite easily accessible through the gravel at these sites — though something about the topography must have made these spots distinctly fertile in water.

Wells in relation to the walled city. Based on R.C. Hope, The Legendary Lore of Holy Wells of England (1893)

  • A series of wells run more or less along the later course of the canal. Stockwell, also called Plato’s well, lay by Hythe Bridge. Old maps show the little watercourse to which it gave rise running into Castle Mill stream.  Stockwell Street, now Worcester Street and Walton St, took its name from this well; Worcester College grounds now embrace its site. It’s said that Stockwell Street was once much used by travellers coming by the old paths over Port Meadow towards the city. It’s hard to tell what the land/water configuration was here before the building of the railway, but ditches along by the railway suggest more or less canalised water along this route. North of Plato’s well lay Boumann’s or Aristotle’s Well, along what’s now Aristotle Lane; also Walton Well, in Walton Well Road. Renaming the wells in honour of Greek philosophers seems to have been an early modern joke. As a Wood put it, because ‘frequently in the Summer Season visited by our Peripatetics’, university men rambling around.
  • The old north wall of the city was once flanked by a moat or ditch, probably incorporating natural water sources, which survived the ditch as fishponds. Along this line lay St Mary’s well, by Smith gate (now Catte st). In the late nineteenth century, Hurst, in his Oxford Topography, noted that a stream of water still ran across Catte st.
  • Along and around the city’s eastern wall was once a marshy area; Hurst says it was once ‘a morass’ and that the subsoil here is ‘peaty’, unlike that found elsewhere in the city. Here lay the holy well, and others clustered around it, such that which one was particularly holy is now uncertain. Other wells along the line of the wall heading south were Crowell, by the corner of modern Holywell st and Long Wall street; and the Postern well.
  • Along the old south wall of the city, Slaying well (by a slaughterhouse), giving it name to Slaying well lane, later Brewer street, running west from St Aldate’s.

Though the topography made these wells possible, they all had their own history of establishment, development and use, elements of which have come down to us. I say more about some of those that were particularly venerated in a forthcoming post on ‘Groundwater: meanings’, but among the others, the history of Crowell seems especially well recorded, if not easy to piece together. It was first mentioned in the thirteenth century, when a cross was said to stand nearby. A culvert running down the ditch in Longwall Street, excavated in 2002, is believed to have carried the water from the spring to the hospital of St John the Baptist, which preceded Magdalen on the site. According to the antiquary Anthony a Wood, it was subject to a series of improvements in the seventeenth century, when Dr Rawlinson built a little house or cot over it, with his arms and the inscription ‘None will hurt this Well that’s wise/ For this hurts none, but cures the Eyes.’ During the Civil Wars, the ‘little house’ (lying as it did near the defensive wall) was knocked down, ‘and the spring suffocated by the town ditch, to which it joined’. Nonetheless, a spring there was turned into a well and surrounded with stone 1651, and 1666, a little seat was set by it. Hurst, writing in the late nineteenth century, recorded that the sluice (presumably to control the flow of water from the spring) had been dug out about twenty years before.


Within the city, by the seventeenth century if not earlier, pumps may have provided the public face of public wells – perhaps reflecting developments in technology which made such urban pumps feasible. A Wood (cited by Hurst) noted that in 1662, a well was dug for a pump at the east end of the Bocherew’ (ie Butcher Row, now Queen St) (the point of the story was not the digging of the well but great old timber posts found by the diggers). Plot mentions pumps at the Cross Inn (ie Golden Cross) near Carfax, one near the Mount in New College Garden and a third at Buckley Hall. His interest was in the mineral content of their waters. It’s not clear if these were accessible for public use. The VCH mentions a tall stone pump with carved heads in the High Street, called the two-faced pump.

In the 1770s, one of the first tasks addressed by the Improvement Commissioners was the removal of the many pumps obstructing the streets.  

Survivor. By St Mary’s, University Church.

My earlier Managing the Waters post came at this topic from a slightly different angle. It gives complementary detail about past and present efforts to organise community water supplies.

Groundwater now

Institutional and domestic wells are not (to my knowledge) a feature of Oxford water provision now, other than on allotments eg Town Furze – though, alongside river water, groundwater continues to play a part in modern water supply.

Statistics that I’ve found relate to the Thames basin, rather than Oxford specifically, don’t agree with each other, and are probably heavily skewed by London patterns, but Thames Water says groundwater supplies 30%  of local water supply to the river’s 70%. The Groundwater Forum rates groundwater’s contribution higher, at 51% (possibly these are breakdowns of different things).

I take it that the switch from wells to piped water was driven by changing estimates of health effects (fear that wells would take in polluted groundwater, in relatively undiluted form), and perhaps also by convenience: a tap might be preferred to a pump. But changes in the availability of water from different sources may also have been a factor in decisions about supply. An account of central London notes that the yield of wells declined notably during the nineteenth century as a result of increasing use, encouraging a turn to rivers. In London, the effect of that switch has been that the groundwater has risen again, sometimes to challenging levels.

Valued as a source of water for use, groundwater is less welcome when presents itself uninvited in the form of floods. In Grandpont, some residents find them homes flooded not from rivers but from groundwater – supercharged by rain and brimming rivers. The role of groundwater in Oxford floods is the subject of a joint project by the Environment Agency and the British Geological Survey, involving

  • Setting up of a monitoring network: … Over 150 groundwater sites and some 50 surface water sites are now routinely monitored across the area, a large number using digital water level recorders on a 15-minute interval…
  • Development of a conceptual model of the groundwater system: … The development of the conceptual understanding has been aided by the creation of a 3D geological model, which is based on the large amount of data for the area in BGS’s borehole geology database.
Aftermath of Boxing Day floods 2020: Christ Church meadows doing its flood-meadow work

For older accounts of springs and wells, I have relied especially upon Hurst’s Oxford Topography, which offers a commentary on an early seventeenth-century account by Leonard Hutten (itself looking back to the previous century); on Robert Plots’ Natural History of Oxfordshire; and on an eighteenth-century edition of Anthony a Wood’s accounts of Oxford, The Antient and Present State – within which it’s not always easy to separate a Wood and his editor’s prose. I’ve also drawn on work by nineteenth-century geologists and engineers, whom I encountered first when looking at Dredging.

I have visited the Maharajah’s well at Ipsden, but don’t have any digital photos of it, so will have to visit again. Wells in the Chilterns are discussed in an article by Pat Preece in the Oxfordshire Local History Association’s Journal, Winter 2000-1.