--- title: 'Water and boiling in the tea classics' url: 'https://howtotea.me/traditions/china/history/pin-shui' markdown: 'https://howtotea.me/traditions/china/history/pin-shui.md' lang: en date: '2026-10-25' description: "Which water Chinese tea writers chose and how they judged the boil: Lu Yu, Zhang Youxin's spring ranking, Qianlong's silver vessel, Japan's pine wind." --- # Water and boiling in the tea classics 品水 Editors · October 25, 2026 · 10 min read From the 8th to the 18th century, Chinese authors argued about two things: which water to use and when it is ready. The famous ranking of springs turned out to be a legend, and the stages of the boil differ from author to author. ![A scholar in a pavilion while a boy boils water for tea on a stove](https://howtotea.me/images/c/o/v/e/r/cover-75d31aca.webp) 品水 Water and boiling pǐn shuǐ Two questions: choice of water, signs of the boil First code: Lu Yu, The Classic of Tea, 8th c. Spring ranking: Zhang Youxin, 9th c. Critic of the ranking: Ouyang Xiu, 1058 Last judge: Qianlong, silver vessel Degrees in the texts: none, a modern reconstruction Chinese tea treatises come back to water more often than to tea. Powder or leaf could be bought, but water had to be found, hauled and watched over a fire. Writers of different centuries worked on two questions. The first is 品水 (pin shui), judging water: which source to use and how to store it. The second is 候湯 (hou tang), watching the boil: how to tell when the water is ready. ## Lu Yu: mountain, river, well In chapter five of The Classic of Tea, Lu Yu sets out a plain ladder. The best water is mountain water, from "milky" springs and stone pools with a slow current. River water should be drawn far from where people live, well water from wells that are drawn from often. Waterfalls and rushing torrents are not to be drunk: he says they cause "neck sickness." In stagnant gorges during the hot season the water turns dangerous and should be let out first. He divides the boil into three stages, also listed briefly under [water for brewing](https://howtotea.me/brewing/basics/water). At the first boil the bubbles look "like fish eyes," there is a faint sound, and you add salt. At the second the water bubbles at the rim "like a spring, in a string of pearls." Now you scoop out a ladleful, stir a vortex with a bamboo stick and drop the powder into the center. At the third, "waves surge." Past that point the water is "old" and not fit to drink. One move in this scheme looks modern. At the third boil you pour the reserved ladleful back into the cauldron to stop the boil and keep the froth. This is the earliest description of checking a boil with cold water. [Tang boiled tea](https://howtotea.me/traditions/china/history/tang-jiancha) covers the rest of the method. ## A ranking doubted by the 11th century A better-known list also carries Lu Yu's name: the twenty best waters, from the Kangwanggu gorge on Mount Lushan to snow water in last place. It survives only in Zhang Youxin's Record of Water for Boiling Tea. Zhang writes that around 814 he found in a temple a note Lu Yu had supposedly dictated to the official Li Jiqing. Dates for the record itself range from 814 to about 825. In 1058 Ouyang Xiu declared the ranking incompatible with The Classic of Tea: Lu Yu forbade waterfalls, yet the list puts them eighth and seventeenth. The editors of the imperial library Siku quanshu agreed. Ouyang Xiu's main argument was that water is simply good or bad, not first or second. The same record tells the anecdote of Nanling. A messenger dipped water from the middle of the Yangtze, spilled some and topped it up from the bank, but Lu Yu tasted the trick and the man confessed. Ouyang Xiu called the story strange and absurd. Tales like this one, or that of the monk who picked out a water among ten jars, are better read as anecdotes than as a tested method. People followed the ranking all the same. In the 9th century the minister Li Deyu ordered Huishan water carried from Changzhou to the capital by post horses, a "water relay." The poet Pi Rixiu mocked it in verse and compared it to lychees for the imperial consort Yang Guifei. Under Huizong the spring became tribute, according to the Wuxi archive "a hundred jars a month," and in the Ming Li Rihua drew up a "contract for delivering the spring": water came by boat once a month, at three fen of silver a jar. ## Song: water from a closed kettle Under the Song, people whisked powder in the bowl, the method known as [diancha](https://howtotea.me/traditions/china/history/song-diancha), and boiled the water in a closed kettle. Hence Cai Xiang's complaint: in a closed vessel the boil cannot be seen, so "watching the boil is the hardest part." He also writes that crab eyes mean the water has already boiled too long. Whisked powder is fussy: underboiled, the froth floats up, overboiled, the tea sinks. In his Treatise on Tea of the Daguan era Huizong looks for water that is "clear, light, sweet and fresh." He admits that this is a gift of nature and rare, and advises taking what is available: a clean mountain spring, then a well that is drawn from often. He rejects rivers for their "fishy smell and silt." Overboiled water he revives as Lu Yu did: add a little fresh water and hold it on the fire for a moment. Su Shi listens to the kettle in verse: first the crab eyes pass, then fish eyes are born, and soon the wind in the pines will sing. His line "living water must be boiled over a living fire" is quoted to this day. Luo Daqing, in 1252 according to Steven Owyoung, describes the boil by ear alone: chirring, cicadas, a thousand carts. By his rule the kettle comes off between the second and third boil, and one waits for silence. ## Ming: leaf wants mature water Under the Ming, tea became leaf tea and water came under new rules. In 1554 Tian Yiheng wrote an entire book on water, Small Notes on Cooking with Spring Water. He holds that source water is always heavy, and the best especially so. Snow, which Zhang Youxin put last, he praises as "especially good for tea." Rain will do if it is gentle and seasonal; storm water, rotten water and water off the eaves must not be used. From Tian Yiheng also comes the advice to put white pebbles in the jar for the road: they "nourish the flavor and settle the cloudiness." Xu Cigui (Cha shu, 1597) describes [storage](https://howtotea.me/brewing/basics/storage): an old clay vat, not a wooden one, the mouth sealed with bamboo leaves and clay, and water drawn with a porcelain ladle. The fire should be hardwood charcoal burned red so smoke does not reach the water. He names tin the best kettle metal and writes off a copper pot as unfit. Today's [kettles](https://howtotea.me/teaware/utensils/kettles) and [heat sources](https://howtotea.me/teaware/utensils/heat-sources) do the same jobs with other materials. Zhang Yuan (Cha lu, around 1595) explains why Ming rules differ from Song ones. Powder floats at once and therefore wants "young" water. Whole leaf needs fully "mature" water, and in his list every early sign counts as immature water: shrimp eyes, crab eyes, fish eyes, pearls. He calls water ready after the hard boil and the silence, when steam rises in a straight jet. For the rest of Ming brewing, see [Ming leaf tea](https://howtotea.me/traditions/china/history/ming-paocha). ## A muddle of stages Set the authors side by side and the order of stages changes. Fish eyes are the very beginning for Lu Yu, but for Su Shi they come after crab eyes. For adding the tea, Lu Yu takes the second boil, Cai Xiang wants water before the crab eyes, Xu Cigui a little after them, and Zhang Yuan only after a full boil. Part of the contradiction comes down to the form of the tea. A popular myth credits Cai Xiang with five stages and temperatures from 70 to 100 °C. His Cha lu has one sentence on the subject: crab eyes are overboiled water. The five-step series with "shrimp eyes" appears with Zhang Yuan, and there all of these stages are immature. The classical texts give no degrees. Measurements point to physics. According to a 2014 Japanese review, when water is heated from 30 to 90 °C, bubbles of dissolved air grow on the bottom, and from 50 °C some of them rise. Steam bubbles form on the bottom only at 94–95 °C, reach the surface at 96 °C, and ripples begin above 98 °C. The early "eyes," then, are not boiling. The Wenshan group's reconstruction put crab eyes at 90–92 °C, while popular charts say about 80 °C. The authors of the trial themselves note that the result depends on vessel, fire and altitude. What happens inside the water itself is a question of [water chemistry](https://howtotea.me/science/water-and-extraction/water-chemistry). ## Qianlong and the silver vessel Under the Qing, the argument over whose water was best turned into numbers. Emperor Qianlong weighed water from different springs in one silver dou. He took Yuquan, the "Jade Spring" near Beijing, as the standard: its dou weighed one liang. Zhenzhu in Jinan came out heavier by 2 li, Huishan and Hupao in Hangzhou by 4 li, and snow water lighter by 3 li. The spread is a fraction of a liang, about one percent between best and worst. Qianlong reasons that water is valued for taste, meaning sweetness, and for essence, meaning lightness, so he names the lightest "first under Heaven." He calls the southern rankings the southerners' partiality for their own. By his own account, palace servants took the measurements on his trips. The date of the inscription is unclear: Steven Owyoung gives 1784, the imperial catalog of inscriptions Huangchao tongzhi gives 1751. The Wuxi museum also reports a trick of "washing water with water": on the road the Yuquan supply was mixed in a vat with local water and the top layer taken, on the belief that light water floats. ## An echo in Japan From the Chinese wind in the pines, Japan went its own way. A verse by the Southern Song monk Xutang Zhiyu about dipping water and listening to the wind in the pines is linked to the Japanese 松風 (matsukaze). In the [chanoyu](https://howtotea.me/traditions/japan-and-korea/chanoyu) kettle, iron plates are fixed under the base, and steam in the gap makes a characteristic hum. Schools count "six sounds of the kettle" and name them differently. The Nanpōroku splits the states of water into five steps and rates the wind in the pines best. But the text is considered a forgery of the Genroku era: Tachibana Jissan copied it out in 1690. It is disputed enough that it is best used with care on questions of tradition. Japan also has its Hundred Famous Waters (meisui hyakusen). The Environment Agency compiled it in 1985 and added another hundred in 2008. The ministry notes that inclusion on the list does not guarantee the water is fit to drink. In Kyoto the well Samegai was famous: by tradition Murata Juko made tea for Ashikaga Yoshimasa with its water, and later Takeno Joo, Sen no Rikyu and Oda Uraku used it. The well was filled in when Horikawa Street was widened after the Second World War. ## Common questions Which water is best according to the classics? There is no single answer. Lu Yu ranks mountain water above river and well water. Emperor Huizong prizes water that is clear, light, sweet and fresh. Tian Yiheng holds that a good spring is heavy, while Qianlong weighed water and looked for the lightest. They agree on one point: the water must be clean and not stagnant. Do crab eyes really mean 80 °C? The classical texts give no degrees. The numbers come from modern reconstructions and they disagree: in one amateur trial crab eyes appeared at 90–92 °C, while English-language blogs say about 80 °C. The vessel, the fire and the altitude all change the result. Why is water overboiled for Cai Xiang but ready for Zhang Yuan? They drank different teas. Song powder floats at once and needs young water. The whole leaf of the Ming era, says Zhang Yuan, needs fully mature water, after a hard boil. --- ## Navigation - Parent: [History of Chinese tea](https://howtotea.me/traditions/china/history.md) - Previous: [Ming and Qing: the shift to steeped leaf](https://howtotea.me/traditions/china/history/ming-paocha.md)