Tweed Water Spaniel: The Border Dog Inside Every Golden Retriever

What the Tweed Water Spaniel was, and why it still matters in India

The Tweed Water Spaniel was a liver-coloured curly-coated water dog of the River Tweed basin, extinct since the nineteenth century, and it matters to an Indian reader for one reason: a bitch of this breed is the female half of the 1868 mating that founded the Golden Retriever.

That is the whole case for reading about an animal nobody can buy. The Golden Retriever is among the most-bought imported breeds in India, and three Indian price pages read on one day quote it anywhere between Rs 20,000 and more than Rs 1,50,000 PawSpace, Dog Price List in India Wrapinfur, Dog Price List in India 2026. The dog those pages are selling has a documented parent that no longer exists, and the documentation is unusually good.

The American Kennel Club states the descent in its own words. Dudley Marjoribanks, the first Lord Tweedmouth, "kept scrupulous records of breedings effected to create an ideal gundog for use at his Guisachan estate in the Highlands, Inverness-shire, Scotland", and "he crossed his Yellow Retriever with a breed that is now extinct, the Tweed Water Spaniel. Irish Setter and Bloodhound were also added to the mix". That is a national registry naming an extinct breed as a founder, which is rarer than it sounds.

Where it lived, and why the map is not the map Wikipedia prints

The dogs belonged to the River Tweed basin on the border between England and Scotland, and Ria Horter's survey of vanished breeds puts it plainly: they "lived almost exclusively in the border region between England and Scotland, the Tweed basin". The two place names attached to the breed sit on opposite banks. Norham is in Northumberland, in England. Ladykirk, which gives the breed its second name, is in Berwickshire, in Scotland. Wikipedia's infobox gives the country as England, and that is a simplification of a dog that lived on both sides of a river that is itself the border.

Three names, one dog

• Tweed Water Spaniel, the form the registries and encyclopedias use today.
• Tweed Spaniel, the form the nineteenth-century sources actually use, including the 1882 technical description.
• Ladykirk Spaniel, after the small town on the Tweed, and the form that ties the dog to a specific place rather than to a river system.

The four documents that describe it, and what each one is worth

Four sources carry almost everything known about this breed, and they were all opened for this page rather than quoted from each other. They disagree about what the animal even was.

The sentence that should govern everything else written about this breed

Horter, having gone looking, reports: "In spite of investigations I have not been able to trace a true depiction or a reliable eye witness." She then adds, about the engraving usually printed as a Tweed Spaniel, that "according to tradition, this engraving should be of a Tweed Spaniel". According to tradition. That is the evidentiary base for every picture of this dog you will ever see.

The colour claim needs the same care. Horter records that the coat "was liver coloured and curly, but liver coloured should be interpreted widely. In those days sandy, fawn and brownish were also used to describe liver-coloured dogs". So a nineteenth-century writer calling a dog liver was covering a colour range, not naming a shade.

The most-quoted description of this breed is attributed to the wrong man

Wikipedia attributes the Tweed Water Spaniel's physical description and its bloodhound speculation to Hugh Dalziel. The passage was written by J. S. Skidmore, in a chapter bylined to him, and it does not describe a separate breed at all.

Dalziel's *British Dogs* is a compilation. Its title page reads "By HUGH DALZIEL, Corsincon, assisted by eminent fanciers", and individual chapters carry individual bylines: the Norfolk Spaniel chapter immediately before this one is "By Corsincon", which is Dalziel's own pen name, while Chapter XXIX, The Irish Water Spaniel, is "By J. S. Skidmore". The Tweed passage is inside Skidmore's chapter.

Here is what Skidmore actually wrote, at pages 149 and 150 of the 1882 edition, which are the page numbers Wikipedia gives:

"When I first commenced to keep Irish water spaniels, many years ago, there were three strains, or rather varieties, one was known as the Tweed spaniel, having its origin in the neighbourhood of the river of that name. They were very light liver colour, so close in curl as to give me the idea that they had originally been a cross from a smooth-haired dog; they were long in tail, ears heavy in flesh and hard like a hound's, but only slightly feathered, fore legs feathered behind, hind legs smooth, head conical, lips more pendulous than M'Carthy's strain."

The part that changes the verdict, not just the byline

Skidmore is not describing a breed. He is listing three strains, or rather varieties, of the Irish Water Spaniel, and the Tweed spaniel is one of them; the other two are the North and the McCarthy strains, which he goes on to describe in the same paragraph. His comparison point is M'Carthy's strain, the founding Irish Water Spaniel line. So the single best physical description of this animal treats it as a regional variety of a living breed, and that is a substantive finding rather than a citation quibble.

The bloodhound remark is about one bitch

Wikipedia says instances of liver-coloured offspring tan below the knees "were noted" and that the author speculated this "may have been due to Bloodhound ancestry in one of the parents of the litters". Skidmore's own sentence is narrower and better: "The one I owned, which was considered to be one of the best of them, I bred from twice, and in each litter several of the puppies were liver and tan, being tanned from the knees downward and under the tail. I came to the conclusion that she, at any rate, had been crossed with the bloodhound."

She, at any rate. The man who owned the dog limited his inference to that dog, and the encyclopedia generalised it into a claim about the breed's parentage. That is the commonest way a hedge disappears.

The 1816 quotation, and the word that goes missing

Wikipedia quotes Richard Lawrence's 1816 sentence about the Newfoundland cross and drops one word from it. The word is experimental, and it is the difference between a practice and an attempt.

The Internet Archive scan of the 1816 edition runs to about 1,994,000 characters and carries a real text layer, so the passage can be read rather than trusted. It says:

"Along the rocky shores and dreadful declivities beyond the junction of the Tweed with the sea of Berwick, water-dogs have derived an addition of strength, from the experimental introduction of a cross with the Newfoundland dog, which has rendered them completely adequate to the arduous difficulties and diurnal perils in which they are systematically engaged."

Wikipedia's version reads "from the introduction of a cross with the Newfoundland dog". One adjective, and the sense moves. Lawrence is describing something people were trying on that coast, not a settled foundation cross, and a reader building an ancestry argument on the shorter version is building on a firmer claim than the source makes.

Which Newfoundland he meant

Almost certainly not the giant breed that carries the name today. Modern writers read Lawrence's Newfoundland dog as the St John's water dog, the smaller working dog of the same island, which is itself extinct and which also stands behind the Labrador Retriever. That reading is a modern inference rather than something Lawrence says, and it should be labelled as one.

What the passage does establish

Two things, and they are both useful. First, that working water dogs at the Tweed mouth were a recognised local population in 1816, half a century before Guisachan. Second, that the people keeping them were salmon netters and cliff fowlers, not sportsmen: Lawrence's cliffs are the ones "so favorable to the propagation of soland geese, sea-gulls, and wild fowl of every description". This was a commercial working dog on a hard coast, which is a different animal from a gentleman's gundog and explains the toughness Lord Tweedmouth was buying into.

Bonbeno's two-condition test: a real dog that was never a registered breed

To count as a breed on this site an animal needs a written breed standard and a studbook. The Tweed Water Spaniel had neither, and unlike some extinct dogs it never came close, because it died before British breed registration existed.

The verdict, and why it is not the same as the turnspit's

The Tweed Water Spaniel is a local type that ran out of time. It is not the same case as the turnspit dog, which was a job description that several unrelated dogs satisfied at once. This was a real regional population with a place, a colour, a coat and a job, whose people knew it by name and bought and sold it, and which simply had no register to enter when registers arrived.

That distinction matters for how a reader should treat descent claims. A type can genuinely found a breed, and this one demonstrably did, because Lord Tweedmouth wrote the mating down. What it cannot do is supply a pedigree, and there is no sense in which a modern dog can be part Tweed Water Spaniel in any documented proportion.

The private studbook is the interesting half

Two sources describe it and they disagree by five years. Horter says "from 1835 to 1890 Lord Tweedmouth kept a studbook in which he recorded not only the names of his dogs, but also their origins". The American Kennel Club says "for the 50 years between 1840 and 1890". Both cannot be right about the start year, and neither is obviously more authoritative, so this page prints both rather than picking one.

Horter adds a second document: "There is another document, dating after 1884 and written in his Lordship's own hand, in which Sir Dudley had noted down the litters of Belle and Tweed". A hand-written litter record for a named Tweed Spaniel bitch is as close to a pedigree as this breed ever came.

The 1868 mating, named dog by named dog

One mating in 1868 at Guisachan in Inverness-shire produced four yellow bitches, and every Golden Retriever alive descends from it. The Tweed Water Spaniel is the dam's side of it.

The record is unusually specific for a nineteenth-century kennel, and the names survive.

Why so few crosses matter more than the number suggests

Two or three matings is a very narrow founding event. It means the Tweed Water Spaniel's contribution to the modern Golden Retriever entered through a handful of individual animals in a single decade, and it means the sentence a reader should carry away is not that Goldens are part Tweed Spaniel in some proportion. It is that a specific bitch called Belle, from a specific river, is in the pedigree of every Golden Retriever in every Indian home.

The theory that gets this backwards

One published account holds that the Tweed Water Spaniel was named after Lord Tweedmouth because it was developed on his estate, and was simply the original name for what became the Golden Retriever. Three things contradict that. The breed is named for the river, and Ladykirk is a town on it. Lawrence describes the Tweed-mouth water dogs in 1816, five years before Lord Tweedmouth was born and forty-five before he took the title. And Guisachan is in Inverness-shire, two hundred kilometres from the Tweed.

The Curly-Coated Retriever claim, tested against the registries

The Tweed Water Spaniel is widely listed among the Curly-Coated Retriever's founding breeds, and that claim does not appear in either registry document that would carry it.

Wikipedia's Legacy section names Poodles, Wetterhouns, Barbets, Irish Water Spaniels, Flat-coated Retrievers, the St John's water dog, the Large Rough Water Dog and the Tweed Water Spaniel as contributors to the Curly-Coated Retriever, and sources the whole list to a magazine feature.

So it was checked where it would show up.

How to read that table, because it is easy to over-read

The two FCI rows are not evidence against the Curly claim. Neither standard has a BRIEF HISTORICAL SUMMARY section, so neither mentions any ancestor of any kind, and a document that says nothing about history cannot be quoted as denying a history. That distinction gets lost constantly in breed writing and it is worth stating.

The AKC row is evidence, and it is the only real evidence here. The Curly-Coated Retriever page's history field exists, is readable, and contains no Tweed. The identical search on the Golden Retriever page in the same session returns eleven hits, which proves the instrument works American Kennel Club. So the honest verdict is that the Golden Retriever descent is registry-stated and the Curly-Coated descent is club and magazine tradition that no registry repeats.

When it disappeared, and why no source will give you a year

No source dates the extinction of the Tweed Water Spaniel, and this page will not invent one. The documents place living dogs from 1816 to at least the 1870s and then stop.

The 1882 tense change is the closest thing to a date

Skidmore is writing in 1882 about a period he places many years ago, and he uses the past tense throughout: there were three strains, they were very light liver colour. Two paragraphs later he switches to the present tense for the Irish strains that survived: "In Ireland, too, there exists two totally distinct varieties, which are now known as the North and the McCarthy strains". One writer, one paragraph, two tenses, and the Tweed is in the past one. That is the firmest dating signal this breed produces and no ranking page has used it.

Why the extinction has no clean edge

Because the dogs were absorbed rather than lost. Horter's general observation about vanished breeds applies exactly here: "the disappearance of some members of a group with the same characteristics does not happen overnight ... some simply disappeared; others live on as the foundation of one or more purebred dogs of today". A local working population that stops being bred as itself and starts being bred into something else does not have a last specimen. It has a last mention.

What it looked like, and the one detail that is a real accuracy test

The Tweed Water Spaniel was a medium-sized liver-brown dog with a close curly coat, a long tail, a conical head and heavy hound-like drop ears, and the single most checkable detail in the whole description is that its forelegs were feathered and its hind legs were not.

That comes from the one person on record who owned one and described it. Skidmore's list, in order: very light liver colour, so close in curl that he suspected an ancestral smooth-coated cross, long in tail, ears heavy in flesh and hard like a hound's, but only slightly feathered, fore legs feathered behind, hind legs smooth, head conical, and lips more pendulous than the founding Irish strain.

Why the feathering detail is the useful one

Because it is asymmetric, specific and useless as a guess. Anyone inventing a description of a curly water dog produces feathering everywhere or nowhere. Fore legs feathered behind and hind legs smooth is the kind of detail somebody writes down because they were looking at the animal, and it is the single best test of whether an image or a description of this breed came from the source or from imagination.

The two things no source gives

No height and no weight. Not one of the four sources publishes a measurement, and the two period authorities who describe overall type say only that it was "a small type of Retriever". If you meet a spec card giving this breed a height range in centimetres, that range has no document behind it, and the rest of that card should be read the same way.

Registry recognition today: none anywhere, checked with controls

No kennel club has ever recognised the Tweed Water Spaniel, and four registers were read in full with in-document controls rather than assumed.

The one registry document that does name it

The American Kennel Club's Golden Retriever history names the Tweed Water Spaniel directly, as an extinct breed used in the founding crosses. That is the correct way to record a breed like this: not as a member of the register, but as a named extinct ancestor inside the history of a breed that is. It is also the only registry text anywhere that says the name out loud.

What that means for the Kennel Club of India

India is a full member of the FCI, represented by the Kennel Club of India in Chennai, and the KCI issues pedigrees against FCI nomenclature. There is no FCI number for the Tweed Water Spaniel and there never will be, so no KCI pedigree can exist for it. What a KCI pedigree can name is the Golden Retriever, FCI 111, and that is what an Indian buyer should be checking.

Tweed Water Spaniel price in India: there is none, and the search proves it

No Tweed Water Spaniel price exists anywhere, because no Tweed Water Spaniel exists. What the Indian price query actually returns is a Golden Retriever price page, four generic price lists and a piece of merchandise.

The result set was pulled from Yahoo India on 9 September 2026 for tweed water spaniel price in india, which returned About 16,900 search results. Here is what those results actually are.

The control, because a search result is only evidence when the instrument is proved

The query labrador retriever price in india was run through the same endpoint in the same session and returned ten genuine Indian Labrador price pages under About 16,200 search results. So the fetch works, the India geo works, and the Tweed result set is a real reading of what India's search index holds rather than a substitution.

The merchandise result is the one to understand

A product listing carrying a breed name looks exactly like price data to a scraper and to a reader skimming a results page. This project has recorded the same pattern before, where the only rupee figure attached to an extinct British breed anywhere in India was the price of a downloadable poster. Before quoting any rupee figure for a rare breed, confirm the listing is for a live animal. On this query, none of them is.

What the living descendant actually costs in India, from three sources on one day

Three Indian sources read on 9 September 2026 give the Golden Retriever at Rs 20,000 to Rs 60,000, Rs 20,000 to Rs 90,000 and Rs 30,000 to more than Rs 1,50,000. The ceiling disagreement is ninety thousand rupees, and that spread is the most useful thing on this page for a buyer.

This is the section an Indian reader came for, even if they arrived by way of an extinct spaniel.

The figures were tested for a disguised foreign price, and they pass

Indian breed pages routinely publish a dollar price run through one exchange rate, so every figure above was divided by 84.38, by 74 and by 80. None divides cleanly at 84.38 or 74. At a flat 80 the quotients are 250, 750, 375, 1,125 and 1,875, and the flat-80 test is structurally weak because any rupee figure ending in three zeros divides cleanly by 80 by construction. Quotients of 1,875 and 1,125 are not prices a shop would print. So these are rupee-native figures, disagreeing with each other in rupees, which is a different and more honest problem than a converted list.

What is actually driving the ninety thousand rupee gap

1. Registration. A KCI-registered puppy with pedigree documentation sits at the top of every band, and an unregistered pet-quality puppy sits at the bottom. The document is most of the price.
2. City. The one source that publishes city bands puts Mumbai highest and Hyderabad lowest, with a Rs 15,000 gap at the floor between them.
3. Who is publishing. A site that also brokers or advertises puppy sales has a commercial reason to quote a wide band with a low entry point, and a site quoting Rs 20,000 for a metro Golden Retriever is describing an undocumented litter.
4. What is not driving it. Health screening. None of the three sources ties price to a hip or elbow score, which is the thing that should be driving it.

Monthly cost of ownership in India for a 25 to 34 kg retriever

A retriever of this size costs roughly Rs 5,000 to Rs 12,000 a month to keep properly in an Indian metro once food, parasite control, grooming and a share of annual veterinary care are counted, and the published Indian estimates for the same dog range from Rs 500 to Rs 20,000, which tells you how loosely the word cost is used.

Start with the disagreement, because it is instructive. Of the three Indian sources that publish a monthly figure for this size of dog, one says Rs 500 to Rs 6,500, one says Rs 5,000 to Rs 9,000 and one says Rs 8,000 to Rs 20,000 Wrapinfur, Dog Price List in India 2026 PawSpace, Dog Price List in India. A floor of Rs 500 a month for a 30 kg dog is not a cost estimate, it is a marketing figure.

Why no monthly food figure appears above

Because Indian retail food prices move fast enough that a number published here would be wrong within weeks, and this project has already watched a headline large-breed bag price change under it inside days. The durable part is the arithmetic rather than the price: take grams per day from the pack for a 30 kg dog, multiply by thirty, divide by the pack weight, multiply by today's price. Do it at the point of purchase.

The spread on one identical vaccine, which is actionable

The same 9-in-1 vaccine is Rs 899 at a private clinic chain and Rs 600 at a Goa government veterinary hospital inclusive of the certificate, while a vet-at-home service prices it from Rs 1,999 plus a Rs 699 visit fee. That is a 3.3 times spread on one drug, and it tells an Indian owner that the price is a function of the delivery model rather than of the medicine.

Could a Tweed Water Spaniel handle the Indian climate? The heat tolerance answer

Bonbeno rates this breed 1 of 5 for heat tolerance, and the rating is an inference from origin climate rather than a measurement, because nothing about this dog was measured. India's mildest coastal month is 9.38 degrees hotter than the hottest month its home estuary produces.

Two Border points were queried from NASA POWER by explicit latitude and longitude, with the grid-cell elevation echoed back and checked, because POWER returns a cell rather than a point.

Berwick-upon-Tweed, at the river mouth, returned a cell elevation of 26.15 m against a real town elevation near 20 m, an error of about 6 metres. That is the most accurate cell in this whole batch of six breeds and it makes Berwick the baseline. Norham, the village an 1833 sporting magazine called long famous for a brown water spaniel, returned 101.95 m against a real 30 m, an error of about 72 metres, so that cell is reading higher ground than the river haugh and is reported as such.

The comparison that settles it

Berwick's hottest month averages 14.71 C. Chennai's coolest month averages 24.09 C. India's mildest large coastal city, at its gentlest, is 9.38 degrees hotter than the warmest month the dog's home estuary produces. Delhi's June mean of 35.88 C is more than twenty-one degrees above it.

A small detail that shows the data is real

Berwick peaks in August, not July, and Norham eight miles inland peaks in July. That is the maritime lag: the sea warms slowly and holds heat, so a coastal point runs a month behind an inland one. It is the kind of thing a copied table never gets right, and it is a cheap way to check whether a climate column was measured or assembled.

Humidity, wet bulb, and why the damp-air argument is the wrong one

Indian humidity is not what would have troubled a Border water dog, because the Tweed estuary is wetter than every Indian city in this comparison. What decides is heat and humidity together, which is what a wet-bulb temperature measures.

The instinct to write that Indian humidity would finish a British water dog is a reasoning error, and this project has made it before. Berwick-upon-Tweed runs 85.75 percent mean annual relative humidity and Norham 87.49 percent. Chennai runs 72.39 percent, Mumbai 67.39 and Bengaluru 67.94 NASA POWER 2001-2020 climatology…. On damp air alone, the dog's home coast wins comfortably.

So the honest instrument is wet bulb, which combines the two into the number that governs whether an animal can shed heat by evaporation at all. Every figure below is computed by the Stull method from POWER's own T2M and RH2M for 2001 to 2020, and the method is named because two different wet-bulb sources are in use across this site and they disagree by up to 1.24 degrees.

Read the gap, not the individual numbers

Delhi's peak wet bulb is 26.50 C and Berwick's is 13.12 C. That is a gap of over thirteen degrees in the variable that governs evaporative cooling, and it is wider than the air-temperature comparison suggests because Delhi's July is both hot and wet. A dog that copes with a dry Delhi May can be in trouble in a wetter Delhi July at a lower air temperature.

What the UK veterinary data says about which dogs actually collapse

The largest study of canine heat-related illness comes from the breed's own country. Hall and colleagues, working with 905,543 UK dogs under primary veterinary care in 2016, recorded 1,259 events from 1,222 dogs; of the 879 events with a recorded trigger, 652 were exertional, which is the 74.2 percent the paper prints, against 113 environmental and 46 from vehicular confinement.

The companion paper in *Scientific Reports*, a different paper with a different case definition, models breeds against the Labrador Retriever and gives the Golden Retriever odds of 2.65 (1.40 to 5.01). That is the living descendant, significantly above a Labrador in a temperate country. In India it is the number that should govern when you walk the dog.

Temperament: one word survives, and it is the right one

Almost nothing is recorded about how a Tweed Water Spaniel behaved, and the one trait that survives in the record is the one Lord Tweedmouth was buying: biddability in a retrieving dog.

Wikipedia says the breed was known for intelligence, courage and sporting ability, and sources that to a coffee-table breed book rather than to a period account. There is no period behavioural description of these dogs at all, and the best modern survey could trace no reliable eyewitness. So this page will not give you a temperament score for the animal itself.

What can be said, from the purpose rather than the dog

Horter records the aim of the Guisachan cross plainly: "there was a reason to breed the Tweed Spaniel to Sir Dudley's yellow Retrievers. The aim was to breed a dog that could find and retrieve game from great distances and out of water." That is a specification, and it implies three things about the dogs being crossed in: they marked and remembered a fall, they entered cold water without being made to, and they gave the bird up at the end.

The continuity to the living breed is documented on both ends. FCI Standard 111 opens the Golden Retriever's temperament clause with the word biddable, and lists intelligence and natural working ability. A single word travelling from a nineteenth-century estate specification into a current international breed standard is about as much behavioural continuity as any extinct dog can claim.

What an Indian household should take from that

Biddable is not the same as easy. A biddable retriever wants a job and will invent one if you do not supply it, and the commonest complaint about Golden Retrievers in Indian apartments is not aggression or noise, it is a bored adolescent dog dismantling the house between eight and eighteen months. The historical record and the modern standard agree on what this line of dog is for, and it is not ornament.

Training: the one quality the whole record agrees on, and where it went

Nobody can train a Tweed Water Spaniel, because none has existed for well over a century. What the record does establish is why Lord Tweedmouth wanted the dog in the first place, and that quality is the single thing that survived the breed into the dog an Indian reader can actually buy.

Biddability is the reason the 1868 cross happened at all. Lord Tweedmouth was assembling a gundog that would find and bring back shot game over rough ground and cold water, and he kept written records of the breedings for fifty years while he did it. A dog that will not take direction is useless for that job, so the trait was being selected for deliberately, generation by generation, in a private kennel. That is the closest thing to a training record this breed has, and it is indirect.

The descendant carries it forward in writing. FCI Standard 111 opens the Golden Retriever's temperament clause with kindly, friendly and confident, and files the breed in Group 8 with a working trial attached, which means the registry expects the dog to be trained to a standard and tested on it. A working-trial qualification is not a personality description. It is a claim that the breed can be taught a job and examined on it.

What this means for an Indian owner who wants to train the living breed

Start the training before the heat does. An Indian retriever puppy has two usable windows a day for most of the year, roughly before eight in the morning and after six in the evening, and a training session that runs into the middle of a Chennai afternoon is a heat-risk session rather than a productive one. Hall and colleagues found that 652 of the 879 UK heat-related illness events with a recorded trigger were exertional, which is 74.2 percent, and exertional means the dog was doing something rather than being left somewhere. Training is doing something.

Indian law also fixes when the work can begin. Rule 8(1)(a) of the Prevention of Cruelty to Animals (Dog Breeding and Marketing) Rules, 2017 requires that pups less than eight weeks in age are not sold, so a lawfully obtained puppy arrives at eight weeks or later and the household socialisation window opens there rather than at six. A seller offering a six-week-old retriever puppy is offering an unlawful sale, and the readiness of that seller to break one rule is information about the rest of the operation.

The claim to be careful with

Trainability is a breed-level tendency and not a guarantee about one animal. No behavioural study of the Tweed Water Spaniel exists, no survey of it was ever run, and PubMed returns zero records for every name form the breed has, against controls of 24, 30, 128 and 126 in the same session. So the honest statement is that the historical record shows the trait was selected for, the modern registry standard still describes it, and neither of those is a measurement.

Exercise: what a Border water dog actually did, and what that costs in an Indian city

A Tweed Water Spaniel worked a cold northern estuary in water, which is the highest-effort job in the gundog world and the least transferable to India. The exercise requirement of the living descendant is roughly two real outings a day, and the Indian constraint on it is temperature rather than distance.

The dogs worked where the Tweed meets the sea. Lawrence's 1816 account places water dogs along the rocky shores and dreadful declivities beyond the junction of the Tweed with the sea at Berwick, engaged in what he calls arduous difficulties and diurnal perils. Strip the period prose and what is left is a dog that entered cold moving water repeatedly, daily, as a job. A dog bred to swim in an estuary at nine degrees is a dog built to make heat, not to shed it.

The temperature gap is the whole argument. Berwick-upon-Tweed averages 9.69 C across the year and 14.71 C in its warmest month, from a NASA POWER cell whose elevation came back at 26.15 m against a real town elevation near 20 m, so the cell is genuinely reading the town. Chennai's coolest month averages 24.09 C. India's mildest coastal month is 9.38 degrees hotter than the hottest month this dog's home country produces, and that gap does not close anywhere in the plains.

The two-outing rule, and the hours that make it safe

Two outings a day, both outside the heat, is the working answer for a large retriever in India. That means roughly 45 to 60 minutes of real activity split across an early-morning and a late-evening session, with the middle of the day spent indoors. The rule is not about the dog being lazy. Hall and colleagues modelled heat-related illness against the Labrador Retriever as the comparator and put the Golden Retriever at odds of 2.65, with a confidence interval of 1.40 to 5.01, which clears one and is therefore a real elevation rather than noise. A breed already carrying 2.65 times a Labrador's odds does not need an afternoon walk.

Swimming is the exercise that transfers best and the one Indian cities supply least. A retriever that can swim will take more work at a lower thermal cost than the same dog walking, because the water carries the heat away. Most Indian apartment dwellers have no access to safe water, so the realistic substitute is retrieving games on grass at either end of the day, in shade, with water available throughout.

What happens without it

An under-exercised large retriever in an Indian flat becomes destructive first and overweight second, and the second one matters more. Bodyweight at or above the breed and sex mean carried odds of 1.42 for heat-related illness in the Scientific Reports cohort, and dogs over 50 kg carried odds of 3.42. So the exercise a household skips in April is paid for in May, through a heavier dog with a worse thermal margin.

Grooming a close curly liver coat in Indian humidity, and the ears that decide it

The Tweed Water Spaniel's coat was close and curly and liver-coloured, which is the single worst coat description to read on an Indian breed page, because a curl that traps water against skin is a curl that traps humidity against skin. No grooming account of the breed survives, so everything below is stated as an inference from the coat description and from the living breeds that carry the same coat.

The 1882 technical description is the closest thing to a coat standard this dog ever had, and it was not written about a breed. Skidmore files the Tweed spaniel as one of three strains or varieties of the Irish Water Spaniel and describes the forelegs as feathered behind with the hind legs smooth, which is an unusually specific detail and the best accuracy test available for any picture claiming to show this animal. Wikipedia attributes that passage to Hugh Dalziel; Dalziel edited the volume and Skidmore wrote the chapter.

A working water dog's coat is waterproof by construction, and waterproof is the problem in India. The oils and the tight curl that keep an estuary out also keep a Mumbai monsoon in, so the coat dries from the outside while the skin underneath stays damp. That is the condition that produces hot spots and pyoderma in Indian retriever populations, and it is why the grooming answer here is drying rather than trimming.

The three things that actually matter, in order

1. Dry the dog properly, and never air dry. Overseas breed clubs routinely tell owners to towel a water dog and let it air dry, which is correct in a northern climate and wrong in an Indian one. Towel to damp, then finish with a cool-setting dryer or a fan, and pay particular attention to the armpits, the groin and behind the ears.
2. Dry inside the ears every single time. A drop-eared dog with a curly coat has no airflow into the canal, and water left there is the commonest route to otitis externa in Indian gundogs. Plug the ears before a bath, and dry them after a swim as carefully as after a wash.
3. Brush to the skin, not across the top. A close curly coat mats at the base while the surface still looks tidy, so a brush that only touches the outer curl reports a coat that is fine when it is not.

What grooming a dog of this coat type costs in India

Budget roughly Rs 900 to Rs 1,900 a session at a chain clinic, before add-ons. Vetic's Delhi price list gives bathing from Rs 934, a haircut from Rs 1,359, complete bath and styling from Rs 1,869 and add-ons from Rs 99. Those are one chain's promotional floors on one city page rather than a market average, and the same chain quotes different prices on different city pages on the same day, so treat the figure as a starting point and price your own neighbourhood. A monsoon-season dog of this coat type realistically needs a professional session every four to six weeks and home drying after every wetting in between.

Health issues and life expectancy: nothing is known, and here is what replaced it

No health data exists for the Tweed Water Spaniel, in any form, and that has to be said three times rather than filled in once. There is no studbook to count deaths in, no breed health survey, no screening scheme and no lifespan study, and PubMed returns zero records under every name form the breed has. Every health issue described below belongs to the living dog an Indian reader can actually buy.

The zero is controlled, which is what makes it worth printing. The query '"Tweed Water Spaniel"[tiab]' returns 0, '"Tweed Spaniel"[tiab]' returns 0 and 'Tweed[tiab] AND spaniel[tiab]' returns 0, all three with clean unrewritten translation lines, while four controls run in the same session returned 24, 30, 128 and 126 records. One name form does produce a number: '"Ladykirk Spaniel"' returns 1,468 records, and that is PubMed silently collapsing the phrase to spaniel across all fields. A four-figure count on a breed nobody has studied is a rewrite, not a literature.

There is no gene pool to study, and that is the actual finding

The breed did not merely die out, it was absorbed. Its gene pool was folded into the retrievers being assembled in the same decades, which is why the AKC can name the Tweed Water Spaniel in the Golden Retriever's own history while no register ever held the Tweed itself. A closed population that ends leaves records behind it. A local stock that gets bred into something else leaves its inheritance inside another breed's studbook, where it can no longer be measured separately. That is the situation here, and it is the reason a health section for this animal has to point somewhere else.

The living breed's two documented risks, with the numbers

Hip dysplasia is the headline, and it has a recent number from a comparable market. Jardim and colleagues, working on Brazilian kennel data published in Veterinary Sciences in August 2025, describe a reported hip dysplasia prevalence in the Golden Retriever of up to 19.6 percent, and estimate heritability for hip score at 0.15 with a posterior standard deviation of 0.08 across a pedigree sample of 1,686 dogs of which 951 were phenotyped. The finding an Indian buyer should read twice is the last one: the genetic trend in hip scores has been stable since 1975 under simple phenotypic selection. Decades of choosing the better-looking hip has not moved the breed.

Heat is the second, and India makes it the first. The Golden Retriever sits at odds of 2.65 for heat-related illness against a Labrador comparator, with an interval of 1.40 to 5.01, and 74.2 percent of the UK events with a recorded trigger were exertional rather than environmental. Both papers describe a British climate. Neither modelled an Indian one, so the numbers are a floor rather than an estimate.

Tick-borne disease, which no overseas breed page will mention

Tick-borne disease is the Indian risk that does not appear in any British or American health survey, and the Indian numbers are large. A survey of 584 dogs in West Bengal found Babesia gibsoni in 76.54 percent and Anaplasma in 55.75 percent of the dogs screened, with 72.9 percent of those 584 screened carrying two or more pathogens. A Kerala survey of 120 dogs found Babesia in 80 of the 120 screened, which is 66.7 percent, with Ehrlichia canis at 5.0 percent and a pre-monsoon March to May peak. Both sampled dogs presented at veterinary clinics rather than a random population, so read them as the burden among dogs tested at clinics and not as a national prevalence.

A long feathered coat on a dog that works in undergrowth is a tick collector, which is why parasite control is a fixed monthly line rather than a seasonal one. NexGard Spectra for a 15 to 30 kg dog runs about Rs 1,103 a month at Supertails' listed price of Rs 3,308 for three tablets, and it covers the tick and flea role and the roundworm and hookworm role in the same tablet. Drontal Plus is the separate deworming route at Rs 125.67 a tablet, dosed one tablet per 10 kg.

Feeding a large retriever in India, and the growth mistake that costs a hip

Feed a large-breed retriever puppy on a large-breed growth diet, measured rather than free-fed, and keep it lean through the first year. That is the one feeding instruction with a mechanism behind it on this page, because growth rate is the part of hip dysplasia an owner controls after the puppy is bought.

Nothing is known about what a Tweed Water Spaniel ate. The dogs belonged to salmon netters and wildfowlers on a nineteenth-century Border estuary and no feeding account of them survives, so this section, like the three before it, describes the living breed. What is worth saying about the historical animal is that it was a working dog in constant cold water, which is a high-energy life, and that nothing about that energy budget transfers to a dog lying under a fan in Hyderabad.

The three rules that matter for a large-breed puppy in India

1. Choose a large-breed growth formula, not an all-breeds one. The calcium and energy density are the reason, and a large-breed puppy fed a generic puppy food grows faster than its skeleton can carry.
2. Measure every meal against the pack's chart, and adjust for the dog rather than the chart. Body condition is the instrument. You want to feel the ribs easily under a thin cover without seeing them.
3. Split the day's ration across two or three meals rather than one. A deep-chested large breed fed one large meal is the classic bloat presentation, and an Indian household eating late often feeds the dog late with it.

Keeping a growing retriever lean is the intervention with the best evidence behind it. The Brazilian pedigree study makes the point from the other direction: hip scores in the breed have not improved genetically since 1975 under phenotypic selection, with heritability at 0.15. A heritability of 0.15 means most of the variation in the outcome is not genetic, which is exactly why growth rate, body weight and exercise load in the first year are where an owner's leverage sits.

What it costs, and where the Indian figures disagree

Indian monthly running-cost figures for this size of dog disagree by a factor of four, and the disagreement is the useful part. PawSpace gives a Golden Retriever monthly cost of Rs 8,000 to Rs 20,000, Wrapinfur gives Rs 5,000 to Rs 9,000 and GoodFurs gives monthly expenses of Rs 500 to Rs 6,500. A floor of Rs 500 a month is not a feeding budget for a 30 kg dog on any food sold in India, and a page printing it is filling a template field rather than costing an animal. Build your own figure from the food you can actually buy in your city and add the parasite and vaccination lines separately.

The first year month by month, and why this timeline is not about a Tweed Water Spaniel

This timeline describes a Golden Retriever puppy in an Indian city, because no Tweed Water Spaniel puppy has been born since the nineteenth century and no account of one survives. It is printed here because it is what a reader searching this breed name is actually about to need.

The legal start point is eight weeks. Rule 8(1)(a) of the 2017 Rules requires that pups less than eight weeks in age are not sold, so month two is where a lawfully obtained puppy joins the household.

The two months that decide the next ten years

Months six to nine carry the most risk and the least visible feedback. Skeletal growth is fastest, the dog looks robust, and the damage from over-exercise or over-feeding in that window does not present until the dog is two or three. The Brazilian hip study is the reason to take this seriously: with heritability at 0.15 and no genetic improvement in the breed's hip scores since 1975, the environmental half of the equation is where the outcome is actually decided.

Registration is the month-twelve job most Indian owners skip. Delhi's Municipal Corporation charges Rs 500 for a new pet dog registration and Rs 500 to renew, with a late fee of Rs 100 a month, and its own counter records 18,958 certificates issued since 2021. Greater Chennai Corporation licenses a pet at Rs 50, issued 9,115 licences in 2024, and runs six municipal pet clinics that provide treatment and anti-rabies vaccination free of cost. A free rabies vaccination from a municipal clinic is the best-value line item available to an Indian dog owner and almost no breed page mentions it.

Head to head: what an Indian buyer is actually being offered instead

Nobody can sell you a Tweed Water Spaniel, so the useful comparison is between the living retrievers and water spaniels available in India and the two breeds this dog is claimed to have founded. One of those two claims survives a check and the other does not.

How to read that table

The Tweed row exists to be crossed out, and the Curly Coated row exists to be corrected. The Curly claim is repeated on breed-club pages and encyclopedias, and it fails a controlled check: the string Tweed appears zero times in the American Kennel Club's own Curly-Coated Retriever history and zero times in FCI Standard 110, while the same search on the same site's Golden Retriever page returns eleven hits in the same session. A zero next to an eleven from the same instrument is a controlled absence rather than a failed search. Neither FCI standard carries a historical summary section at all, which is worth knowing before reading anything into their silence.

The last column is the one that should drive a purchase, and it is the one no ranking page prints. The Tweed link is a story. A hip score is a document, and a seller who cannot produce one for both parents is telling you something.

The word for what the Tweed Water Spaniel was

Unrecognised is the accurate word, and unlisted is not. The Royal Kennel Club's complete Breeds A to Z returns zero for tweed against ten in-document controls, the American Kennel Club's complete breeds index returns zero against nine controls, the FCI letter T index holds ten breeds and none of them is this one, and the FCI's provisional register holds twenty entries and none of them is this one either. Four registers, four controlled absences. This was a local Scottish and Northumbrian working stock that was absorbed before breed registration reached it, and calling it unrecognised states a fact about the registers rather than a grievance about them.

What Indian readers get wrong about the Tweed Water Spaniel, in order of cost

Eight mistakes recur on this breed, and they are ordered here by what each one costs the person making it rather than by how often it appears.

1. Paying for one. There is no Tweed Water Spaniel at any price, anywhere. The first organic result for the Indian price query is a Golden Retriever price page, followed by four generic Indian dog price lists and an Amazon listing for a notebook. Any seller quoting a rupee figure under this name is selling a different animal.
2. Buying a Golden Retriever in April. The breed carries odds of 2.65 for heat-related illness against a Labrador comparator, and a puppy arriving at the start of an Indian summer meets its worst conditions during its least robust months. Time the purchase for October if you can time it at all.
3. Believing the Curly Coated Retriever descent claim. Tweed appears zero times in the AKC's Curly history and zero times in FCI 110, against eleven hits on the same site's Golden page.
4. Attributing the 1882 description to Hugh Dalziel. J. S. Skidmore wrote Chapter XXIX; Dalziel edited the volume and bylined his own contributions Corsincon. More importantly, Skidmore does not treat the Tweed as a separate breed at all.
5. Quoting the Lawrence sentence without the word experimental. The 1816 text says water dogs gained strength from the experimental introduction of a cross with the Newfoundland dog, and dropping that one word turns an attempt into an established practice.
6. Printing a lifespan, a height or a weight for this breed. None was ever published, there is no standard to take them from, and every spec card carrying those fields generated them.
7. Shaving a retriever's coat for the Indian summer. It removes the layer that reflects sun, disrupts airflow at the skin and frequently grows back patchy on a curly coat.
8. Treating an overseas breed club's care advice as climate-neutral. Guidance written for a Border estuary tells you to towel the dog and let it air dry, which in an Indian monsoon is how a drop-eared dog gets otitis.

Adoption and responsible sourcing: what to do instead, and what the law allows

You cannot adopt a Tweed Water Spaniel and you cannot import one, so the practical question is what an Indian household should do with the interest that brought it here. Adopt a retriever or a retriever cross from an Indian rescue, or adopt an Indian pariah dog, and buy from a breeder only after the two checks below.

Indian rescues carry surrendered retrievers constantly, and they carry them for a reason that is relevant to this page. A large, biddable, high-energy dog bought on impulse in a hot city is one of the commonest surrenders in Indian metros, so the same breed this page routes readers toward is already sitting in shelters in every city named on it. An adult dog from a rescue also removes the entire growth-rate gamble described in the feeding section, because the hips have already finished forming and can be assessed as they are.

Why importing is not an option, stated from the instrument

No import authorisation exists for this purpose, and that is a legal fact rather than a practical difficulty. Dogs enter India under ITC(HS) 01061900, classified Restricted, and DGFT Trade Notice 22/2022-23 states in Policy Condition 8 that import of commercial dogs for breeding or any other commercial activities is not permitted, with paragraph 6 adding that any applications for trading or breeding purposes shall be duly rejected. Entry is limited to seven ports, at Delhi, Mumbai, Chennai, Hyderabad, Bengaluru, Kolkata and, since October 2024, Kochi. An extinct breed has no exporting population in any case, so the import question is doubly closed, but the notice is the document to cite rather than the absence of listings. The full route is set out in how to import a dog to India.

The two checks to run on any Indian breeder

1. Confirm the puppy is at least eight weeks old. Rule 8(1)(a) of the 2017 Rules requires that pups less than eight weeks in age are not sold, and a seller offering younger is breaking the rule in front of you.
2. Ask how the parents are related. Second Schedule Part II paragraph 6(3) of the same Rules prohibits in breeding, defined as breeding between relatives of the second to fourth grade within the first four generations, and prohibits incest breeding between parents and children or brother and sister. A breeder who cannot answer the question does not have the pedigree to answer it with.

Register the dog once it is home. Delhi's Municipal Corporation charges Rs 500 for a new pet dog registration, and Greater Chennai Corporation licenses a pet at Rs 50 and runs six municipal clinics that give treatment and anti-rabies vaccination free of cost. Registration is cheap, it is the legal position in both cities, and in Chennai it opens a free vaccination route that no private clinic can match.

What this breed's ending is actually for

The Tweed Water Spaniel is worth a page because of how completely it vanished. A working stock on the Scottish and English border, valuable enough that a peer crossed it into his own kennel and kept records for fifty years, left no standard, no studbook, no photograph and no reliable eyewitness. What survived is inside another breed. That is the argument for writing down what a local population is while it still exists, and it is the argument for taking the Indian pariah dog seriously as a breed rather than as a default.

Compare with similar breeds

Tweed Water Spaniel sits in 3 Bonbeno groups. Each one compares it directly against the breeds it is genuinely an alternative to, on the same figures.