Why Doesn't Ganga Water Rot? The Science Behind India's Strangest River Claim

In a great many Indian kitchens there is a small brass or copper pot pushed to the back of a shelf, behind the spice tins, holding water that nobody intends to drink. It was filled at a ghat — Haridwar, Rishikesh, Varanasi, Prayagraj — on a trip that may have been a pilgrimage or may have been a funeral. It has been sitting there for years.

Every so often somebody takes it down, holds it up to the tube light, and looks. And the water is clear. Not cloudy, not green, not sour. Clear.

That small domestic observation, repeated across a hundred million households, is the root of one of the most persistent claims in Indian public life: that Ganga water does not rot. It shows up in temple discourse and in WhatsApp forwards, in government press releases and in tribunal judgments. It has been offered as proof of divinity, and it has been offered as proof of science, sometimes in the same paragraph.

So it is worth asking the question properly, without either reverence or sneering. Does Ganga water actually resist putrefaction? And if it does — what exactly is doing the work?

The honest answer is more interesting than either camp usually admits. Something real is happening in that pot. It is just not the thing most people think it is, and it is running out of room.

A British bacteriologist and a result he could not explain

The scientific thread begins in 1896, in a laboratory in Agra, with a man who was not looking for a miracle.

Ernest Hanbury Hankin was a British bacteriologist working in colonial India on the problem that mattered most to public health at the time: cholera. Cholera moved along rivers, and it moved along pilgrimages, and the Ganga was both. Hankin's working assumption was the obvious one — that river water thick with human traffic would be an excellent medium for Vibrio cholerae, the bacterium that causes the disease.

It was not. When he added cholera bacteria to water drawn from the Ganga and the Yamuna, the bacteria died. Not slowly. Within hours, populations that thrived happily in distilled water or well water were wiped out.

Hankin then did the thing that makes him a scientist rather than an enthusiast: he tried to kill the effect. He boiled the water. The killing power vanished. He passed the water through a filter fine enough to trap bacteria. The killing power survived.

Those two results, taken together, told him something specific. Whatever was destroying the cholera bacteria was too small to be a bacterium itself, and it was fragile enough to be destroyed by heat. In other words, it was alive, or close enough to alive to be broken. Hankin published the observation and, sensibly, stopped short of naming a cause. He did not have the tools.

He had, without quite knowing it, described a bacteriophage — a virus that infects and destroys bacteria — roughly two decades before Frederick Twort and Félix d'Hérelle independently found them in Europe and gave them the name. An Indian river had produced one of the earliest recorded observations of the most abundant biological entity on Earth, and the observation sat mostly unexamined for a century.

The Three-Layer Filter: how any river cleans itself

Before we can say whether the Ganga is special, we need a model of what "clean" even means for a river. Here is the one worth carrying around. Call it the Three-Layer Filter.

Every flowing river cleans itself through three stacked mechanisms, and they work in a fixed order.

Layer one is dilution — physics. A river carries volume, and volume spreads a pollutant thin. Ten thousand litres of sewage entering a torrent is a different event from the same sewage entering a trickle. This layer is entirely about how much water is actually moving.

Layer two is oxidation — chemistry. Turbulent water pulls oxygen out of the atmosphere at every riffle and rapid. Dissolved oxygen lets microbes break organic waste down into harmless components. This is why fast, shallow, tumbling rivers recover so much faster than slow, deep, sluggish ones. It is also why the standard measure of river pollution is biochemical oxygen demand — BOD — which is really a measure of how much oxygen the waste will steal on its way to being digested.

Layer three is predation — biology. Rivers contain organisms that eat other organisms, including bacteriophages that hunt specific bacteria. This layer does not dilute waste or oxidise it. It kills particular things.

The order matters enormously, and it is the part almost every viral post gets wrong. Layers one and two do the overwhelming bulk of the work in every river on the planet. Layer three is a specialist. And crucially, layer three depends on layers one and two staying intact — phages need water to travel through and hosts to find, and both collapse when flow collapses.

The Ganga's genuine peculiarity is concentrated almost entirely in layer three. Its reputation, however, is usually explained as though layer three were doing everything.

What the research actually found in layer three

After decades of anecdote, the question finally got money. In 2014–15 the Indian government funded a study, reported at around ₹150 crore, to investigate the Ganga's self-purifying properties, drawing in scientists from AIIMS, the IITs and the National Environmental Engineering Research Institute.

Microbiologists at the Institute of Microbial Technology in Chandigarh, working on a project commissioned by the water resources ministry in November 2014, reported finding a range of bacteriophages in Ganga water and linked them to its non-putrefying character. That is the core, defensible finding: the river carries an unusually rich phage population, and phages suppress bacteria.

Supporting work points the same way. A 2009 study by C. S. Nautiyal, published in Current Microbiology, found that Ganga water accelerated the death of a pathogenic strain of E. coli. Research cited in the National Green Tribunal's 2017 Ganga judgment describes work at the National Botanical Research Institute in Lucknow in which Ganga water dosed with five times the normal load of E. coli still cleared it — and in which stored Ganga water, sixteen years old, retained the ability to do so. The tribunal also cited older work by D. S. Bhargava, from 1982, finding that the Ganga drops its BOD faster than comparable rivers, and a Malaria Research Centre observation that mosquitoes did not breed in the river's upper reaches.

More recently, researchers examining the 2025 Maha Kumbh at Prayagraj have argued that bacteriophages may have quietly suppressed infection transmission during a gathering of extraordinary density.

Put together, this is a real body of evidence for a real phenomenon. Ganga water is genuinely hostile to certain bacteria in a way that ordinary river water is not. The pot on the kitchen shelf is not lying.

Why the pot stays clear but the river does not

Here is where the popular version quietly substitutes one claim for another.

"Non-putrefying" means the water does not go foul — it does not develop the bacterial bloom and the smell that stagnant water usually develops. That is a statement about bacteria. It is not a statement about safety, and it is emphatically not a statement about health.

Bacteriophages are the most specific killers in biology. A phage does not attack "germs" as a category; each one attacks a narrow range of bacterial hosts, often a single species or even a single strain, by locking onto a particular molecular structure on the cell surface. That specificity is precisely why phages are so promising as medicine — and precisely why they are useless against most of what actually pollutes the Ganga today.

Phages do nothing to chromium from a tannery. Nothing to nitrates and phosphates from fertiliser runoff. Nothing to pharmaceutical residues, microplastics, arsenic, or the dissolved organic load from untreated sewage. A phage cannot lower BOD. It cannot restore dissolved oxygen. It has no opinion about heavy metals at all.

So the bottled water in your kitchen can be simultaneously clear, non-putrefying, and completely unfit to drink. Those are not contradictory findings. They are answers to different questions.

It is worth remembering how narrow the word "clean" gets when you push on it. The purest water human beings produce is the ultrapure water used to rinse silicon wafers, and its purity is measured in parts per trillion — one of the many reasons building a semiconductor fab is so much harder than it looks. Drinking-water clean, ritually clean, non-putrefying and fab clean are four different standards, and a river can pass one while failing the rest.

The numbers that do not agree with each other

In February 2025, during the Maha Kumbh, the Uttar Pradesh government issued a press release stating that samples collected from five major bathing ghats by a Padma Shri–awarded scientist showed no bacterial growth and no drop in pH, and that Ganga water contains 1,100 types of bacteriophage. Because of this self-purification, the release said, the water remained unpolluted even after 57 crore devotees had bathed in it.

Three weeks earlier, on 3 February 2025, the Central Pollution Control Board had filed a report with the National Green Tribunal covering the same river during the same event. It found high biochemical oxygen demand at Sangam and six other locations, including on the major bathing dates, and faecal coliform levels several times above the permitted standard.

Both statements were made about the same water in the same weeks. They cannot both be describing the whole picture.

The inconsistency runs deeper than one festival. Compare two versions of the same claim that circulate freely: the tribunal's 2017 judgment records the assertion that the Ganga's oxygen levels are twenty-five times higher than other rivers, while the social-media version of the claim says twenty-five per cent more. A factor of a hundred separates those two numbers, and both are quoted as established fact. When a figure can drift that far without anyone objecting, it has stopped functioning as data and started functioning as a slogan.

The 2020 peer-reviewed study of ritual bathing during the 2013 Maha Kumbh is the most useful corrective, because it holds both halves at once. It found that pollutant levels rose sharply and dissolved oxygen fell during the event — and that Ganga water carried more bacteriophages than the Yamuna, which did help control bacterial growth. Its conclusion was neither triumphant nor dismissive: the self-cleansing capacity is real, and it is being overwhelmed.

Flow is the quiet hero of this story

Ask a river scientist what keeps a river clean and you will rarely hear "viruses" first. You will hear "flow."

That is the finding buried in the tribunal's own reasoning: rivers possess self-purification capacity mainly because of flow speed, which maintains oxygen supply and breaks down organic waste. Any river with genuine flow purifies itself. The Ganga is somewhat better at it than most. But the mechanism is layers one and two, and the Ganga's famous layer three rides on top of them.

Which is exactly the problem. Between Rishikesh and Prayagraj, the river's flow is drawn down for hydropower, irrigation and drinking supply until, in the dry months, it can slow to something close to a standstill in places. A group of secretaries constituted in June 2014 to examine Ganga rejuvenation noted pollution signals appearing even in the upper reaches, where oxygen levels should be at their best. The then Union water resources minister put the situation more bluntly still, acknowledging that the river runs pristine only until roughly Haridwar and then takes on the character of a drain as it crosses the plains.

The sentence that captures it: the river's flow stops, but the sewage does not.

This is the same physical logic that governs the Indian water year more broadly — a subcontinent whose entire hydrology is a seasonal gift that has to be caught and released carefully, which is why India is not a desert despite sitting at desert latitudes. Take too much out of the system between monsoons and you are not just short of water. You have switched off the river's ability to clean itself.

What the Ganga might still be worth, scientifically

There is a version of this story that is genuinely exciting, and it has nothing to do with proving anyone's faith right.

Antibiotic resistance is one of the defining medical problems of the century, and India sits close to its centre. Phage therapy — using bacteriophages as precision weapons against resistant bacteria — is one of the more credible responses, and it needs enormous libraries of well-characterised phages to draw from. Researchers have already isolated phages from the Ganges that break down biofilms of drug-resistant Klebsiella pneumoniae in laboratory and zebrafish models.

Seen that way, the river is not a medicine. It is a repository. Its value is the diversity it holds, and that diversity is a function of a specific, unrepeatable combination: glacial origin, sediment chemistry, the particular mineral load it picks up crossing the Himalaya, the nutrients it carries into the plains. As one environmental scientist put it in a 2025 fact-check of the viral claims, no two rivers have identical chemical compositions, which is why the Ganga's microbial profile differs from the Yamuna's — and why we cannot yet say whether other rivers hold similar treasure, because the comparative studies have not been done.

We are, in other words, in the position of people who have found an unusual library and are still arguing about whether the building is holy, while the shelves are being emptied.

India already takes microbial inventories seriously in other contexts — enough that planetary protection scientists worry about Earth microbes hitching rides to the Moon. The microbial community of a river that a quarter of a billion people live alongside deserves at least the same rigour.

Is a river ever really beyond saving?

It is tempting to read all this as an obituary. It is not, quite.

Layer three — the phages, the strange microbial inheritance — is the hardest layer to rebuild and the one that nobody knows how to engineer. Layers one and two are, bluntly, policy. Flow is a decision about how much water we abstract and when. Dissolved oxygen is a consequence of flow. Sewage load is a function of how many treatment plants actually run.

Conservation history is full of recoveries that looked impossible until the specific bottleneck was identified and attacked. The kākāpō went from 51 individuals to more than 300 not through hope but through relentless, unglamorous intervention on the one or two variables that actually mattered.

Rivers are the same. And we are, for the first time, able to watch the whole basin at once rather than sampling it at a few ghats — which is much of the point of putting an imaging satellite in a fixed position over India and staring at the same landscape all day. Continuous monitoring turns "the river seems worse this year" into a measurable, arguable, fixable claim.

What the pot on the shelf is actually telling you

So: does Ganga water rot?

Less readily than most water, yes. There is a genuine biological reason, documented since 1896 and partially characterised since. The water in your grandmother's brass pot is clear because of something real, and it is a small piece of scientific history that the first hint of bacteriophages anywhere in the world came out of an Indian river.

But that clarity is a narrow property, not a general one. It says the bacteria did not bloom. It does not say the water is safe, or healing, or unpolluted. And the property depends on a river that keeps moving — which is the one thing we have been steadily taking away.

The most useful thing the Three-Layer Filter gives you is a way to hear these claims properly. When someone tells you the Ganga cleans itself, the right question is not "is that true?" It is "which layer are you talking about, and is that layer still working?"

Because the river has been answering that question honestly for a long time. Water that has been standing for sixteen years in a sealed pot can still kill E. coli. Water flowing past Prayagraj in February could not keep its oxygen up for six weeks. Both facts belong to the same river. Only one of them is our doing.

Frequently asked questions

Does Ganga water really never rot?
It resists putrefaction better than most water, and that resistance is linked to an unusually rich population of bacteriophages — viruses that destroy bacteria. But "does not go foul" is not the same as "is clean." Stored Ganga water can stay clear while remaining unsafe to drink.

Is Ganga water safe to drink?
Not in most of its course. The Central Pollution Control Board's own monitoring has repeatedly found biochemical oxygen demand and faecal coliform levels well above permitted standards at major locations, including during the 2025 Maha Kumbh. Bacteriophages do not remove sewage load, heavy metals, nitrates or pharmaceutical residues.

Who first discovered that Ganga water kills bacteria?
The British bacteriologist Ernest Hanbury Hankin, in 1896, observed that Ganga and Yamuna water destroyed Vibrio cholerae, and that the effect survived fine filtration but was destroyed by boiling. He described the behaviour of bacteriophages about twenty years before they were named.

Do bacteriophages make the Ganga self-cleaning?
Only partly. Most self-purification in any river comes from flow — dilution and oxygenation. Phages add a third, highly specific layer that kills particular bacteria. When flow is reduced by abstraction and damming, the first two layers weaken, and the third cannot compensate.

Sources

• The Week / First Check — "Does Ganga water have self-cleaning and self-healing properties?" (February 2025)
• Down To Earth — "Rivers' natural flow is what keeps them clean" (February 2025)
• PubMed — "Self-cleansing properties of Ganga during mass ritualistic bathing on Maha-Kumbh" (2020)
• PMC — "Phages from Ganges River curtail biofilms and planktonic growth of drug-resistant Klebsiella pneumoniae"
• The Indian Express — IMTECH microbiologists on bacteriophages in Ganga water