Older accounts call these formations stone stalagmites that rose from the sea floor and were then worn down by wind and water. Almost every part of that is the wrong way round. Stalagmites are built up inside caves by dripping water; these are karst pinnacles, carved out of solid limestone by water dissolving it away. The rock did form on a sea floor, but it was uplifted and then cut into, not pushed up as pillars. The error is forgivable in one respect — the Chinese word for these columns literally means stalagmite — but it inverts the whole process. Those accounts also claim the Stone Forest is one of the seven natural wonders of the world, which no official body recognises, while omitting that it has been a UNESCO World Heritage site since 2007.
What these pillars actually are
A stalagmite is a cave formation: a mound built up on a cave floor by carbonate precipitating out of water dripping from above. It is depositional — material arrives and accumulates.
The Stone Forest is the opposite kind of landform. Its pillars are karst pinnacles, and they are dissolutional — they are what is left when water removes the rock around them. The bedrock is late Palaeozoic shallow marine carbonate, laid down in the Devonian, Carboniferous and Permian; the Major Stone Forest specifically stands on Permian bioclastic limestone. That rock was uplifted out of the sea, and the pillars were then cut out of it along a well-developed network of joints and two regional faults.
The shaping happened in two distinct phases, and both are needed to explain what a visitor sees. First, while the limestone was still buried under soil and sediment, groundwater dissolved it from below along those joints — which is why the lower parts of the columns carry broad horizontal grooves and rounded, sometimes mushroom-like bases. Then the cover stripped away and the tops were exposed to direct rainfall, which cut the sharp vertical flutes that give the pillars their bark-like texture. Pagoda-shaped columns come from dissolution working along bedding planes. Rounded below, fluted above: two processes, two periods, one pillar.
As for how the mistake arose, the Chinese term for these columns is shisun, which translates literally as "stalagmite" and is used loosely for them in Chinese. Chinese official material in English still describes "stalagmite-like stones". So this is a translation artefact rather than an invention — but it describes the wrong process, and the right one is more interesting.
A World Heritage site since 2007
The Stone Forest is part of South China Karst, a serial World Heritage property inscribed under criteria (vii) and (viii). The Shilin Karst cluster was one of three in the first phase of 2007, alongside Libo in Guizhou and Wulong in Chongqing; a second phase in 2014 added Guilin, Shibing, Jinfoshan and Huanjiang. The whole property now runs to seven clusters across four provinces and municipalities, with 97,125 hectares of core area and 176,228 of buffer.
What UNESCO says about Shilin in particular is worth quoting properly, because it is a stronger claim than the one older accounts make up. Its stone forests are described as superlative natural phenomena and as the world reference site for this type of feature, with a wider range of pinnacle shapes than any other pinnacle karst landscape anywhere. Development is described as spanning about 270 million years, from the Permian to the present, and as episodic rather than continuous. Two named stone forests appear in the inscription: Naigu, which stands on dolomitic limestone, and Suyishan, which rises out of a lake.
One inconsistency to handle carefully rather than smooth over: the Committee's decision describes the Shilin cluster as having two components, while the Chinese management plan describes a core zone in four separate patches — Naigu, Wangchengshan and Qingshui Pond, Wenbi Mountain and Suoyishan, and the Major and Minor Stone Forest. Both are official; they are different documents counting different things.
Two further designations, and the one that replaces the myth
The Stone Forest is also number 80 in the First 100 Geological Heritage Sites of the International Union of Geological Sciences, announced in October 2022 — one of only seven Chinese entries on that list. This is a scientifically conferred honour from the world body for geology, and it is precisely the thing the "seven natural wonders" claim was reaching for.
It is a Global Geopark, and has been since 2004, in China's first batch. It became a UNESCO Global Geopark in 2015, when UNESCO created that label — so describing it as a UNESCO Global Geopark since 2004 is anachronistic. The geopark covers 36,570 hectares and holds a population of around 67,800. It is also a national five-A scenic area and a national key scenic area.
Not one of the seven natural wonders
Older accounts say the pillars are "indeed one of the seven natural wonders of the world, even though they might have narrowly missed receiving the official title", which contradicts itself in a single sentence. There is no official title to miss.
UNESCO put this on the record in a statement of 9 July 2007: there is no link whatsoever between its World Heritage programme and the "New 7 Wonders" campaign, a private initiative launched in 2000 which UNESCO was invited to support several times and declined. It noted that the campaign's results reflect only the views of people with internet access. No intergovernmental body maintains a list of seven natural wonders at all.
The traditional Chinese epithet, however, is real and officially used. The Yunnan provincial tourism authority calls the Stone Forest the First Wonder Under Heaven. What fails is the attribution: the famous "Stone Forest Wonderland" inscription at the Major Stone Forest is attributed by the same provincial source to Long Yun, a governor of Yunnan in the Republican period — the 1920s to 1940s — not to the Ming dynasty, and not to "the locals". No Ming-dynasty attestation of the phrase could be traced to any official or scholarly source.
How big it is, and which figure means what
Older accounts give 96,000 acres, which is about 388 square kilometres and matches no official figure exactly. There are at least five, and they measure different things:
The World Heritage core for the Shilin cluster is 120.7 square kilometres, with a buffer of 229.3. The scenic and protected area is 350 square kilometres, a figure given by both the Ministry of Culture and Tourism and the provincial tourism authority. The UNESCO Global Geopark is 365.7 square kilometres. And the wider karst landscape runs to over 1,100 square kilometres.
The 350 figure reconciles cleanly with the management plan's zoning — a core of 120.7, a buffer of 100.21 and a tourist-serving zone of 129.09 — which is the tidiest way to present it. The 96,000 acres in circulation is most likely a rounded conversion of the geopark area.
How tall the pillars are
Not hundreds of feet. The UNESCO geopark description gives the Major Stone Forest's pillars as 20 to 50 metres, roughly 65 to 165 feet. The management plan gives ranges of 5 to 40 metres in one place and 30 to 50 in another; the provincial tourism authority says the tallest free-standing column exceeds 40 metres; and scholarly measurement gives columns of 2 to 34 metres, with stone teeth under 5 metres making up nearly half the total.
So "hundreds of feet" overstates it, though less wildly than it first appears — the tallest columns do pass 150 feet. One figure not to confuse with pillar height: the site itself sits at 1,500 to 2,203 metres above sea level, and those altitudes sometimes get repeated as if they described the stones.
The age, and the two dates that get conflated
"Over 200 million years ago these stalagmites rose from the sea floor" collapses two separate things and invents a third.
The rock was deposited as marine carbonate in the Devonian, Carboniferous and Permian. The pinnacles developed over roughly the following 270 million years, from the Permian onwards, in episodes rather than continuously; the management plan distinguishes four evolutionary types, including a palaeo stone forest, a petrified one, one still developing and one baked by basalt. The forms visible now are dated mainly to the Middle Pleistocene. And there is a buried interval in between: around 50 million years ago a lake laid red clay over the pinnacles, which were later re-exposed — the limestone is overlain by red Palaeogene sandstone and Permian basalt.
Nothing rose from the sea floor as a pillar. The rock formed on a sea floor, was lifted, and the pillars were cut out of it. "Dozens of peaks" is also out by orders of magnitude: this is thousands of columns, in dense clusters.
The sections, and the two that get left out
The provincial tourism authority lists four scenic areas, not three. The Stone Forest Scenic Area comprises the Major and Minor Stone Forests, the latter officially the Ashima Scenic Area. Naigu, also written Heisongyan, is a scenic area in its own right rather than a third section of the first. Then there is the Feilong Waterfall area and Changhu. The geopark's own reporting names four subareas: Major, Minor, Naigu and Changhu, and a national forest park at Guishan sits alongside them.
So older accounts omit both Changhu and the waterfall, and demote Naigu to a subdivision. Spellings vary even between official documents — Naigu and Nagu, Suyishan and Suoyishan — so one form has to be chosen and held.
The waterfall, and a tell-tale conversion
The official English name is Feilong Waterfall, with Dadieshui in brackets, on the Bajiang river; "Dadie" in older accounts garbles the second name. The management plan names it as a scenic area and an entrance but gives no height.
The figure usually quoted is about 88 metres, which could not be traced to an official source. What is clear is that the legacy figure is not a measurement: "almost 300 feet (91 m)" is 300 feet converted into metres, which means the writer rounded first and converted afterwards. That is the same signature as the invented tomb mound on the Xi'an page, and it is worth recognising.
Four things that could not be verified
Older accounts carry several features that no official or scholarly source supports, and they are worth naming so that nobody plans around them.
The Strange Wind Cave, with gales of two or three minutes returning every thirty like clockwork from August to November, is the least supported thing in the legacy text. The cave is named in a zoning table in the 2005 management plan — name only, no description, no mention of wind. It appears nowhere in the World Heritage documents, the geopark material or the geological heritage listing, and the specific claim traces to a specialist-looking reference site that cites an encyclopedia and a travel blog for it, and whose own author could not locate the cave on any map. The underlying physics is plausible, since siphon-driven intermittent springs do displace air, but nothing published confirms it here.
Hongxi Spring and Penfeng Cave appear in no official source at all and are probably garbled romanisations. Zhiyun Cave is real and is named in the management plan, in the buffer zone of the Naigu area — singular, and with no dimensions ever published, so the "720 underground acres" has no basis. And the underwater stalagmites said to be visible in Long Lake are unverified; anything submerged there would be drowned karst pillars, repeating the same error. Underground rivers, by contrast, are well documented: the management plan discusses complex systems of them, windows onto subterranean rivers, and the need to protect their catchments.
On Long Lake itself, the name is a straight translation rather than an error — the official name Changhu means long lake. UNESCO's geopark description calls it a natural karst lake surrounded by green hills, used for traditional festivals and picnics. For a lake with pinnacles rising out of it, the officially documented one is Suyishan.
The wildlife, counted
"Thousands of native animal and plant species" is not right, and the real figures are better for being precise. The management plan records 820 species of seed plant, in 615 genera and 72 families, of which 8 are protected nationally and 20 provincially; an independent assessment gives 889 vascular plant species. For animals it records 185 vertebrates — 42 mammals, 87 birds, 32 reptiles, 12 amphibians and 12 fish. Thousands of species is reachable only by counting invertebrates, which nobody here has.
The Sani, and Ashima
The people of the Stone Forest are the Sani, a branch of the Yi, and UNESCO states that they have lived in what is now the geopark for more than 2,000 years, with a culture closely tied to the karst landscape and geological sites interpreted through the Ashima epic. The World Heritage Committee's own decision says the relationship between the karst and Yi cultural identity requires continued recognition and respect in the site's management, which is an unusually direct instruction to find in an inscription.
The numbers behind that: Yi make up 33.3 per cent of the geopark's population, rising to around 90 per cent in the south, and the core zone holds about 4,000 residents, mostly Sani, at 17 people per square kilometre. The county is Shilin Yi Autonomous County.
Ashima is recognised as national intangible cultural heritage, and the county has its own regulation on the transmission and protection of Ashima culture — a level of formal protection that puts it well beyond local legend. The Ashima Stone Peak is an officially named formation, the Minor Stone Forest is formally the Ashima Scenic Area, and the Ashima figure is floodlit at night. The story, in the provincial heritage account, is of a hard-working Sani girl who pursued love and died for it.
The Torch Festival falls on the twenty-fourth day of the sixth lunar month, confirmed by China's national intangible cultural heritage authority, which describes household torches, a tall pine-branch torch raised at the village entrance, rosin thrown onto the flames, circle dancing, wrestling, bullfighting and torch processions through the fields to drive off crop pests. It is kept by the Naxi, Bai, Lisu, Lahu and Jinuo as well as the Yi. In 2026 the Shilin festival opened on 5 August.
Getting there, and how many are let in
From Kunming there are two sensible routes and they are very different in length. High-speed trains run from Kunming South to Shilin West in just over twenty minutes, with a local bus from the station to the Stone Forest. By road it is about 80 kilometres and roughly an hour and a half. Coaches run from Kunming East bus station, which is on the metro. The frequently repeated "about an hour from Kunming" matches neither.
On capacity, the management plan sets a daily limit for the Major and Minor Stone Forest of below 13,000 visitors, and China has told UNESCO it is strictly controlling numbers across the seven clusters. Visits ran to about a million in 1988, 2.05 million in 2005 and 3.58 million in 2024. No official opening hours could be traced — everything carrying them was commercial — so the practical advice is that capacity is actively managed and the official channel is the one to check.
Nothing is announced for 2027 or 2028. The conservation and management plan for South China Karst ran from 2016 to 2025 and no successor has been published; the most recent state-of-conservation report states that the property has no major restorations, alterations or new construction planned, and records that China abandoned a plan to renovate and widen two roads crossing Shilin Karst and its buffer zone.
Entering China
One date governs everything. British citizens holding standard passports could enter mainland China visa-free for stays of up to thirty days until the end of December 2026, for tourism, business, visiting family or friends, and transit; paid work, study, journalism and longer stays need a visa arranged in advance, and no extension beyond 2026 has been announced. A passport needs at least six months' validity beyond arrival and two blank pages, and place of residence must be registered with the local public security bureau within twenty-four hours, which hotels normally handle at check-in.
A warning about checking this independently: searches for the position after 2026 return a thick layer of machine-generated material confidently predicting an extension. None of it is authoritative. The only confirmed extension to be found anywhere was a bilateral arrangement with Russia, which has no bearing on British travellers.