(Video 5) Positioning the chaire in front of the mizusashi on the utensil mat.

(Video 4) Adding water to the mizusashi.

(Video 3) Positioning the mizusashi on the utensil mat.

(Video 2) Filling the mizu-tsugi.

(video 1) Filling the mizusashi.

Notes on the Practice of Chanoyu — a Short Introduction for People Without Access to a Teacher (7l - Wari-geiko 2: Filling the Mizusashi and Mizu-tsugi, and Setting up the Utensil Mat]

Filling the mizusashi and mizu-tsugi, and arranging the mizusashi and chaire on the utensil mat are likewise pretty straight forward activities, and do not require a lot of explaining to understand.

I. Filling the Mizusashi and Mizu-tsugi.

1) First one mizuya-bishaku full of cold water is used to rinse the mizuashi.

2) Then water is poured in through a mizu-koshi¹ until the mizusashi is 70% full. 

3) Then the mouth should be dried with a towel, and the lid put on², and the outside dried with a towel; and so it is put to the side until the host will take it onto the utensil mat.  (Video 1)

     The same procedure should be followed when filling the mizu-tsugi with water.  (Video 2)

II. Arranging the Mizusashi and Chaire on the Utensil Mat.

1) During the naka-dachi, once the chaire³, mizusashi, and mizu-tsugi have been prepared, the host begins by taking the mizusashi out to the utensil mat.

2) First the mizusashi is moved from near the host’s entrance and onto the temae-za, and from there it is put into place next to the furo.  The host should be sitting directly in front of the mizusashi⁴ when he moves it into position.


◎ As seen in the above sketch, the mizusashi is centered on the left half of the utensil mat (which is not quite the same thing as saying it is centered between the furo, or shiki-ita, and the left edge of the mat, so the host must be careful⁵).  (Video 3)

3) After the mizusashi is properly positioned, the host returns to the preparation room and brings out the mizu-tsugi (the spout of the mizu-tsugi should be oriented so it will be pointing toward the mizusashi when lifted on the temae-za).  The process of moving it onto the temae-za is the same as with the mizusashi.  The host should again sit so that he is directly in front of the mizusashi.

4) First the lid of the mizusashi is removed and leaned against its left side.

5) Then the host turns to the mizu-tsugi, and takes the chakin off of its lid:  if its spout is protected by a hinged metal cover, the host moves this open using the chakin.

6) Picking up the mizu-tsugi by its handle, the host pours water into the mizusashi until it is 90% full.  While pouring, the chakin should be in the right hand, and held beneath the spout, to help control the stream of water.

7) When enough water has been added, the host puts the mizu-tsugi down again, closes the cover of the spout with the chakin, and returns this attention to the mizusashi.

8) He dries the mouth of the mizusashi with the chakin — right side, left side, and front — being careful not to reach over the mouth of the mizusashi (since dust may fall from his sleeve or arm into the water).

9) Then he picks up the lid, cleans it with the chakin — back and front — and then closes the lid. If the mizusashi has projecting ears that may have been splashed with water, he may dry them with the chakin at this time, too.  (Video 4)

10) Then he removes the mizu-tsugi from the room and brings out the chaire

11) First the chaire is moved onto the temae-za.  Then, with the host sitting in the middle of the mat (in the same place he will sit during the temae), he puts the chaire in front of the mizusashi.  (Video 5)

◎ The back side of the chaire should be 2-sun away from the front of the mizusashi at its widest diameter⁶; and the foot of the chaire should be to the right of the kane on which the mizusashi is centered⁷ (as seen in the sketch, below).


12) Before returning to the mizuya to attend to other matters, the host should always pause to make sure that the placement of the chaire is correct.  According to the old teaching⁸, the seam at the front of the shifuku should be aligned with a me [目] of the mat.

◎ Even though there is never anyone to see how the host does these things, you should make an effort to do everything correctly (as explained above), just as if the guess were sitting in their seats and watching.  Chanoyu is a discipline, and this (and everything else you do) is all part of your temae.


¹Mizu-koshi [水漉し]:  a utensil made of aka-sugi [赤杉], red cryptomeria wood secured with thin strips of cherry-bark, and shaped exactly like a mizuya-bishaku, but with the bottom replaced by a piece of sarashi cloth (the same linen cloth that is used to make the chakin), stretched taut by a wooden band, so that the effect resembles an embroidery hoop.


     In the above photo, a mizu-koshi is on the right, and a mizuya-bishaku is on the left.

²This should be done quickly, so that no dust will fall onto the water.  Also, it is better not to cross the mouth of the mizusashi with the right hand:  rather, dry the right side with the towel held in the right hand, and then transfer to towel to the left hand, and so dry that side.

³While there is no hard and fast rule, the chaire is usually filled first, so the host may attend to this when he is not pressed for time (in the early days of chanoyu it was displayed, already filled with tea, on the tana during the sho-za). 

     First the tea should be sifted through a cha-koshi [茶漉し] — of which there are several different types available (all working equally well) — and then gently poured into the chaire through a funnel.  If too much tea is dumped into the funnel the mouth might clog, and tapping it (or plunging the mouth with a chashaku) may result in the tea becoming compacted into lumps which, as katamari, will render the koicha disagreeable.

⁴This is why, on the daisu (and in every other situation), the mizusashi is placed facing straight forward, rather than at an angle as is the furo.

⁵On a kyōma tatami the center of the mat is roughly 1.5 me to the left* of the shiki-ita.

     When the furo is placed on top of the covered mukō-ro, the left side of the shiki-ita† should correspond with the left edge of the heri on the side of the ro.  It is placed so that it is 2-sun back from the front edge of the mukō-ro.

     When Rikyū placed a furo directly on top of the lid of the mukō-ro without a shiki-ita, he oriented the furo just as if the shiki-ita were present (rather than doing something like centering the furo on the lid‡).
*When the furo is placed on the right side of the mat, over the closed mukō-ro.  If the furo is on the left, the center is about 1.5 me to the right of the shiki-ita.  Though this of course depends on the omote, the reed mat used to face the tatami (there is some variation in the width of the me); it is best for the host to count the me the first time, and then see precisely where the center is.

†Assuming that it is an ordinary 9-sun 5-bu square ko-ita, which is by far the most common kind of shiki-ita sold.

‡As some modern tea teachers have suggested. 

     Since placing the furo directly on top of the lid of the closed ro is something done only in an extremely small tearoom, the host must be careful that the furo is not too close to the seat of the guest — often the shōkyaku — who is seated next to the ro.  If the furo is located as if on a ko-ita, the air in front of the guest’s knees will be ho hotter than the ambient temperature of the room; but if it is closer to him than that, the guest may begin to feel uncomfortable (at least during the hottest part of the summer) as the gathering progresses.

⁶So if the mizusashi is narrow at the base, the distance between the chaire and the mizusashi may appear to be greater than 2-sun).

⁷The front side of the chaire's foot is then usually not intruding on the yū-yo, though this is not really important now.  The yū-yo is forbidden territory during the temae.  This is why Rikyū did not like to talk too much about these kinds of things*.

     Also, in consideration of the teaching that is explained in the next note (#8), since the chaire is tied in its shifuku at this time, matching the foot to the kane is not an absloute (the way it is when the shifuku has been removed).
*Displaying the chaire directly on the mat in front of the mizusashi is an extremely wabi way to do things.  Before Jōō came up with the idea for futatsu-kazari [二つ飾] (which means the display of the furo, on a shiki-ita, and the mizusashi on the mat beside it — placing the chaire close to the mizusashi means it is counted together with the mizusashi, rather than considered a separate element in the display), the chaire was always displayed on a tana (or else carried out by hand at the beginning of the temae):  it was only placed directly on the mat during the service of tea.

⁸”When arranging the utensils together [on the utensil mat], take a moment to pause and look:  the seams [of their* shifuku] should be oriented so they are placed on the me of the mat” [置合わせ心を付けて見るぞ可し、袋の縫目疊目に置け].  Poem 100, from the Hundred Poems of Chanoyu.
*While we commonly display only the chaire in a shifuku today, in Jōō’s period the temmoku (or other chawan, such as the large Shukō chawan) was occasionally displayed on the utensil mat in one at the beginning of the temae, too.

     In Rikyū’s densho he addresses the case of the display of the temmoku in a shifuku, and implies that this is something that should be done only rarely.  However these things were written long after Jōō’s death, and after specially made small chawan 4-sun in diameter had become common. 

     Also, in his densho Rikyū usually refers to the display of the temmoku in a shifuku in the context of serving tea to a nobleman (that period’s equivalent to the modern kijin-date [貴人点] class of temae — for people acquainted with the traditional terminology).  What the host does when serving tea to his friends, however, is an entirely different matter, and displaying the chawan in a shifuku was often used as a way to indicate the special reverence felt for the bowl, due to its historical associations.


     Furthermore — and lastly — the non-temmoku chawan most usually displayed in its shifuku during the first half of the sixteenth century had been the Shukō-chawan (the same kind of bowl is seen in the photo above).  This bowl was lost in the fire at the Honnōji when Nobunaga committed seppuku in the summer of 1582, and with it the practice of displaying the chawan in its shifuku at the beginning of the go-za.

(Video 2) Emptying and cleaning the kama.

(Video 1) Filling the kama.

Notes on the Practice of Chanoyu — a Short Introduction for People Without Access to a Teacher (7k - Wari-geiko 2: Preparing the Kama, and Cleaning the Kama at the End of the Gathering)

Filling the kama is pretty straight forward, and does not require a lot of theorizing or explanation.

     By rights, the water for chanoyu should be drawn from a natural well, and this should be done at dawn.  Dawn is the time of day when the sky first lightens, up to several hours before sunrise.  The water at this time was preferred since it is the cleanest — since the well will not have been disturbed since sometime the previous afternoon, any sediment will have settled to the bottom, and the water in the well will have risen to its highest level, thus keeping the concentration of mineral contaminants to a minimum.  For this reason, such water was traditionally preferred for any purpose when especially clean water was needed, such as the preparation of medicines.  As has been mentioned before, this water is known as sei-ka-sui [井華水], which translates as “Flower-of-the-Well Water”

    Traditionally the host prepared a pair of kiji-tsurube, a couple of short lengths of rope, and a bamboo shoulder-pole for this exercise.  Placing the tsurube on the curbstone of the well, the host would lower the well-bucket¹ down into the water of the well and draw it out gently (so as to disturb the water in the well as little as possible).  Then he would pour a little water into the tsurube, to rinse them out.  Then he would fill both of the tsurube to about 70% of their capacity, and put the lids on. Then they would be attached to both ends of the shoulder-pole with the two lengths of rope, and so carried to the mizuya².

     If the host was just bringing the water to the mizuya, the kiji-tsurube were left as they were.  If, however, the host was planning to use one of the tsurube as the mizusashi³, he was supposed to pass a long paper tape through the hole in the crosspiece and wrap it around that tsurube, gluing the two ends together and to the side of the tsurube⁴, sometimes going so far as to sign the ends of the tape or apply his name-seal (to prove that the water had not been tampered with).  The tape was cut using a small knife immediately before opening the kiji-tsurube during the temae.  This practice continued until sometime after 1582 when it began to fall into disuse.  From this time on the tsurube was supposed to be “locked” by putting something on top of the movable half-lid (at the beginning of the sho-za the ceramic kōgō was used to lock the lid, and after the sumi-temae it was replaced by the hishaku), but in the Edo period even this was forgotten and the tsurube was used as it was⁵.

     In the old days, when the water was simply being brought to the mizuya, the host usually filled the mizu-kame [水甕], the large (usually ceramic⁶) storage-jar that is kept in the mizuya for this purpose, and put the rest of the water aside.

     After the kama has been washed inside and out, it should be filled⁷, and then the lid should be closed fully (so no dust can get inside).  When the host is using the ro⁸, the kama should be filled at dawn (according to Rikyū’s san-tan san-ro [三炭三爐] teaching⁹); but, during the furo season, the kama is washed and filled just before the guests are expected to arrive, so it will still be damp when they enter the tearoom for the sho-za.  After dealing with the kama, the mizusashi would be rinsed, and filled to approximately 70% of its volume.  And then, after rinsing, the mizu-tsugi should likewise be filled until the water is about half-way up the opening of the spout on the inside.

     Then the water remaining in the second tsurube was poured into the mizu-kame, to be kept until needed, later in the day¹⁰.

     If you are using tap water, it might still be good to draw it at least several hours before the gathering and store it in a large jar or bucket in your preparation area, so it will be close to room-temperature when used.  Otherwise it may cause water to condense on the outside of the mizusashi and the mizu-tsugi, and make a mess of the utensil mat just when the host has no time to be mopping up the drips.

◎ I understand that people who have never studied chanoyu in Japan (or with a Japanese teacher, at least one of a certain age) may find these digressions irrelevant, if not down right annoying.  Unfortunately, this is how much of what the student needs to know — important teachings that we might call background information — is, and has always been, communicated during study.  It is as important (or maybe more important) than the actual mechanics of setting up the room and performing a temae.

I. Filling the Kama.

1) The kama should never be touched with the bare hands¹¹.  It is placed on the kama-sue¹², and its lid is removed and placed somewhere safe (such as on the shelf of the mizuya).

2) A mizuya-bishaku of cold water is poured into the kama.  The outside of the kama should also be wetted with more water, as needed.

3) Holding the kama with towels between the metal and the host’s hands, the kama is rotated, and then the water is discarded into the sunoko.

4) Placing the kama mouth-downward on the sunoko the host rinses the bottom of the kama.

5) Righting the kama again, the host fills it with the necessary amount of water.  The water should be poured through a mizu-koshi so any dust will be strained out.

6) Then the lid is closed fully and the kama moved away.  (Video 1)

II. Emptying the Kama (at the End of the Gathering).

1)The host prepares the mizuya for the kama by filling the chakin-darai half full of cold water, and placing the kama-sue out on the sunoko.

2) The kama is brought to the mizuya and placed on the kama-sue.

3) Then the host returns to the utensil mat and removes the fire from the furo.  Some of the embers should be placed in the gan-ro¹³ in the mizuya, since it will be used to dry the kama after it is cleaned.

4) Once these things are done the host removes the lid from the kama, dries it, and puts it on a futaoki on the shelf.

5) Then, with towels between his hands and the side of the kama, the hot water is poured into the chakin-darai until it is full. 

6) The rest of the hot water may be discarded into the sunoko., and the kama placed — with the mouth facing downward — on top of the kama-sue.

7) The host takes the kiri-wara from its peg and dips it into the chakin-darai, and then scrubs the bottom of the kama

◎ The host must be careful, since while some pressure is necessary (to remove the oily residue from the incense), too much pressure will damage the finish on the kama.

8) Using the hishaku¹⁴, water from the chakin-darai is poured over the bottom of the kama to rinse the residue away, and then the kama is turned over.

6) After inserting the kan, the kama is removed.  (Video 2)

◎ The kama must be dried over heat as soon as it is emptied.  The best heat-source is several burning embers¹⁵ located 2-sun or so below the bottom of the kama.  It can be dried either in the gan-ro in the mizuya (if the host has one), or the kama may be returned to the furo and dried there.  (In the latter case the host should have left several embers in the furo, since if there is insufficient heat the kama will not dry completely, and rust may develop that will ruin the kama.¹⁶) 


¹Despite what some authors have written, the kiji-tsurube are not “well buckets”.  They are used to transfer the water from the well to the mizuya.  Well-buckets are usually somewhat larger versions of the ordinary wood-and-bamboo hand-bucket that the host uses to refill the tsukubai.

²As has been described previously, the mizuya in Rikyū’s period was little more than a roofed-over veranda attached to the the back side of the ko-yashiki, with a floor made of split lengths of bamboo installed one bu or so apart (like a large version of the sunoko — so any spilled water would drain through without pooling, and dry quickly), and was usually exposed on two or three sides   Hence the person carrying the water simply walked up to the edge of the veranda and lowered the tsurube onto its floor.

³Because the water the kiji-tsurube contains is the purest water that can be used for chanoyu*, since ancient times the kiji-tsurube has been considered to be a meibutsu mizusashi.  As such it was frequently placed on the daisu in the early period for the most formal temae.
*Each time the water is transferred into a new container, it naturally becomes less pure.  Since the kiji-tsurube received its water directly from the well-bucket (which, itself, is too large and inelegant to be used as a mizusashi), this is as pure as the water for chanoyu gets.  Pouring it from the tsurube into the mizu-kame, and from there into a bronze or celadon mizusashi, no matter how fine and beautiful, degrades the water.  Once this was understood (and the need to show off ones collection of meibutsu utensils no longer necessary to impress the guests — as it had been in the earliest days of chanoyu in Japan), the tsurube came to be used as a mizusashi on the daisu whenever the host wished to showcase the purity of the water.

◎ Like the mentsū, the kiji-tsurube was originally used as a mizusashi only one time.  (This is why the host had no qualms about marking it with ink or his name-stamp when sealing it with a paper tape.)

⁴This practice, which is of ancient origins, was to certify* that the water in the tsurube had been drawn at dawn, and had remained untouched until the gathering.  When Jōō invented the cha-kai [茶會]†, he made a point of also applying such a tape to the lid of the kama, to show that no matter when the gathering began, the kama had been prepared at dawn‡.


*In imitation of the practice where water drawn at dawn by the attendants serving the famous wells in the city would stack up tsurube of this water for people to take and use for medicinal preparations.  These tsurube were sealed with a paper tape on which was written the date, and the tsurube itself was stamped with the name of the Shintō shrine that guarded the well (since the tsurube were the shrine;s property and had to be returned).

†While chanoyu had been practiced in Japan for more than a hundred years before Jōō’s lifetime (and the drinking of matcha as usucha known for considerably longer), tea had always been served as part of the refreshments offered to the guests during some other sort of function (such as a poetry gathering, or a gathering organized to appreciate incense).  Jōō was the first to arrange gatherings only to enjoy tea drinking, and the basic format of sho-za (where the fire is laid and the guests served food and kashi while waiting for the kama to begin to boil) go-za (where first koicha, and then usucha are drunk) that is still observed today was wholly his idea.  (Prior to him the tearoom was prepared quietly beforehand while the guests were in another part of the house; and the guests were ushered in only to drink tea, after which they moved on to another room to resume the activity for which the gathering had been called).

‡This was one reason why only high-quality kama were supposed to be used for chanoyu — since low-quality iron will naturally rust and contaminate the water between when it was filled at dawn, and whenever the gathering began.

     Ashiya kama, from Kyūshū, were originally made with Korean iron (which contains molybdenum and other metallic contaminants that inhibit or retard the formation of orange iron-oxide [Fe₂O₃]).  By the sixteenth century (as a consequence of the lapse in trade with the continent following the Ming invasions of the mid-fifteenth century), the Ashiya foundries were using mixed Korean and Japanese iron (which, while inferior to the original kama, resulted in pieces that were still superior — so far as preventing the formation of orange rust — to those case with pure Japanese iron).  Temmyō kama (made in Sano-Temmyō, in the modern Tochigi Prefecture, north of Tokyō), were made using a high quality pure Japanese iron, hence they were traditionally considered to be inferior to Ashiya kama.

⁵In the Edo period an echo of the application of a paper tape (to seal the kiji-tsurube) still was heard from time to time, but what had been done in the days of Jōō and Rikyū was not clearly understood, and so often subject to misinformed interpretation.


     Some chajin confused this “shime-kazari" [飾り or 締め飾り], which means “display of the seal” (see the illustration accompanying note #4, above) with the Shintō institution known as shime-kazari [注連飾] (paper streamers hung from a rice-straw rope to indicate ritual purity), and as a result many modern schools hang a shime-kazari on the kiji-tsurube when displaying the water — which usually means the host is using water brought from some famous well, in a practice called mei-sui-date [名水立, or 名水点].

⁶In his mizuya-dōko Rikyū seems to have used a wooden hand-bucket for this purpose, but that may have been simply because the space available in the dōko is very limited.  A ceramic jar with a square unpainted wooden lid (usually just a flat board, approximately the size of a shiki-ita) is much more common, especially in an ordinary mizuya.

⁷The kama is filled “up to the bottom of the kantsuki.”  The reason is because the area at the top of the kan-tsuki bears the weight of the kama when it is picked up, thus the host must try to keep the iron there as rust-free as possible.  By filling it no higher than the bottom of the kan-tsuki, the host will achieve this aim.

     Some of the modern schools teach that the smaller kama used with the furo should be filled higher (to the middle, or even the top, of the kan-tsuki).

     But if we look at Rikyū’s small unryū-gama (as an indicator of his preferences), we find that three mizuya-bishaku of water fill this kama precisely to the bottom of the kan-tsuki, which accords with the rule.

⁸While Jōō preferred the dichotomy of using the furo during the warm months (which began in the Third Lunar Month, and continued to the end of the Ninth Lunar Month*), and the ro during the cold season (from the Tenth Lunar Month to the end of the Second Lunar Month†), Rikyū considered the ro to be the greatest expression of wabi, and so decided that it could be used in the wabi setting all year round, if the host wished. 

     From the records of his own gatherings, it seems that Rikyū was actually of two minds with respect to this matter:  sometimes he appears intent on observing the change of seasons, just as Jōō had done; and sometimes it appears that he was inclined to use the ro all year round.  While he is known to have placed the furo directly on top of the closed wooden lid of the mukō-ro (to lower the kama closer to the floor‡), it almost seems that his Mozuno ko-yashiki was designed so that the ro would be used all year round (so that the host would also have an unobstructed view of the tokonoma, just as the guests** had).
*Roughly, from early April until sometime in November.  Because this is based on the Lunar Year, the exact dates always change relative to the Solar Year.  Some years the Lunar Months begin earlier, and in other years they commence several weeks later.  Since calculating these things is difficult (every six years an extra month is intercalated — though not at the same point every time this is done, and can sometimes lead to confusion:  for example, while Rikyū was ordered to commit seppuku on the 28th day of the Second Month of Tenshō 19, that year actually had two first months, so the anniversary of his death was observed on the 28th day of the Third Lunar Month by those people who were familiar with the circumstances…which apparently the Senke was not, since they always observed Rikyū-ki on the 28th of the Second Month; and the date of his death equates to April 21, 1592), it is best to rely on a calendar.

†From the middle of November or so, until the end of March.

‡According to Rikyū’s writings, he seems to have understood that the “degree of wabi" for any given gathering was primarily related to the height of the mouth of the kama above the surface of the mats. 

     On the daisu, the furo is raised one sun above the surface of the mats by the ji-ita.  The shiki-ita raises it about 6-bu.  Placing the furo on top of the lid of the mukō-ro (which was level with the surface of the mats) meant that the kama was only the furo's height above the mats.  And when placed in the ro, the mouth of the kama will be only 6- or 7-bu above the surface of the mats — or, perhaps, it may even be below the surface (if it is an uba-guchi kama [姥口釜], or an ordinary kama that was placed deeply in the ro, such as when the ro is used during the summer, to keep the fire as far away as possible from the guests).

     The lower the mouth of the kama, therefore, the more wabi the setting.  In the Mozuno ko-yashiki , during his last year of life, he seems to have frequently used either the ko-arare ubaguchi gama [小霰姥口釜], also known as the hyaku-kai ubaguchi gama [百会姥口釜] or the small unryū-gama suspended on a bamboo ji-zai so that the mouth was 6- or 7-bu below the ro-buchi, and continued to do so even after the season changed in the spring of Tenshō 19.

**Or at least the shōkyaku did.  In this kind of room, the second and third guests would have had to inspect the toko by looking across the shōkyaku's knees.  (The jumping back and forth that the modern schools teach — which is based on the way things are done in the large rooms — should not be done in these extremely small rooms.  When separated from to the utensil mat by less than than a half mat, the guests observe whatever is placed there from their own mat — and, of course, since this room had a mizuya-dōko, nothing would ever be displayed on the utensil mat for them to inspect anyway.)

San-tan san-ro [三炭三炉] refers to Rikyū’s original teaching* regarding the setting up of the ro at dawn


- At dawn the shita-bi [下火] (the “base fire,” usually consisting of three or five small pieces of burning charcoal lighted from the fire preserved overnight in the mizuya) is introduced into the cold ro and a set of charcoal is laid around it just before the host places the just-filled kama into the ro.

- At mid-day (around 1:00 PM) the host enters the tearoom and lifts the kama out of the roShimeshi-bai [湿し灰] (coarse, moist ash,the purpose of which is to keep dust from flying out when the burning charcoal is handled) is sprinkled over the surface of the ash in the ro, and then the charcoal fire is rebuilt (so it will continue to burn until dusk).  Then, using the mizu-tsugi, water is added to refill the kama and it is returned to the ro.

- At dusk the host comes back into the tearoom and removes the kama to the mizuya.  Then he returns and transfers all of the burning charcoal from the ro to a large ceramic bowl of shimeshi-bai (the moist ash keeps the bowl from getting too hot to handle:  this bowl is called a handa hō-roku [半田炮烙]).  Using a large scoop (shaped like an over-sized soup-ladle) he removes the top layer of hot ash and burning embers from the ro. and puts them in the handa hō-roku, too.  After spreading a new layer of shimeshi-bai over the surface of the ash in the ro to replace what he removed, the host selects several of the larger embers from the handa hō-roku and puts them back in the middle of the ro, and lays a new set of charcoal around them.  Then, returning to the mizuya while the fire begins to catch, the host empties the kama and washes it inside and out, fills it with cold water from the mizu-kama, and then takes it back to the ro.

- Around 10:00 PM the host returns to the tearoom and removes the kama and fir from the ro, cleans it, and sprinkles a layer of shimeshi-bai over the surface of the ash in the ro, and leaves it to cool until the following dawn.  Some of the burning charcoal is kept in the mizuya, to start the next day’s fire, and after a careful cleaning the empty kama is put to dry over this small fire.

San-tan [三炭] — which means “three [layings of] charcoal” — refers to the set of charcoal laid at dawn, the rebuilt fire at mid-day, and the new set of charcoal laid at dusk.  San-ro [三炉] — “the three [states of the] ro" — means the cold ro at dawn, the refreshed ro at mid-day, and the hot but completely changed ro at dusk.  The host who understands how the charcoal catches fire and burns when placed in a cold ro†, in a refreshed ro‡, and in a completely changed but still hot ro**, knows how to control the fire, and so will be able to host a gathering successfully no matter what time of the day or night he receives his guests††.  And understanding comes only from experience, from doing this day in and day out, as a kind of Buddhist training.s was probably Rikyu’s greatest contribution to the practice of chanoyu — the legacy that has been completely forgotten by the modern schools today‡‡.
*This teaching is related to the ro, hence could not pre-date Jōō’s time (since the large ro with a kabe ro-dan [壁炉壇] — a deep, mud-plaster liner — was one of his innovations:  prior to Jōō an old rice-cooking pot, with the upper sides above the flange broken off, was sunken into a round hole cut in a board that supported it at the level of the mats, and used to hold the fire; but this arrangement can only support the use of a small fire, about the same size as is laid in the furo).  But since Jōō always started his gatherings with a cold kama, regardless of what time of the day (or night) the cha-kai began, the san-tan san-ro practice could not have originated with him.

†At dawn the ro is cold, perhaps near freezing in an unheated traditional ko-yashiki.  So while the draw of air early in the day is excellent, the ro and ash absorb a lot of heat from the shita-bi, thus the host must arrange the charcoal carefully — since while it begins to burn slowly, the fast draw of cold air will mean that once the fire catches, the charcoal will begin to burn strongly, and can burn out before mid-day if the host does understand — from experience — how to do things.

‡Because the ash was only top-dressed, as it were, heat and the gases of combustion are still present at this time.  Furthermore, because the ambient temperature at mid-day is rarely very cold in the areas of Japan (Sakai, south of Ōsaka, and Hakata, in western Kyūshū — and then Tokyō — are all somewhat warmer than Kyōto:  people generally need to open their windows at mid-day, even in the middle of winter) where chanoyu was mostly practiced, the draw of air into the ro will be slower than when the temperature is near freezing, hence the charcoal catches fire and burns more slowly.

**At dusk, conditions are nearly ideal (which is why the night gathering was usually favored in the early days of chanoyu).  The removal of the fire and top layer of ash means that the source of the gases of combustion will have been removed as well.  And while the ambient temperature in the room falls with the sun, the ro itself is still warm, so it will not sap the heat from the shita-bi.  Thus the newly laid charcoal will catch fire easily, and the cooling air in the room means that the draw will be perfect, insuring that the fire burns uniformly and well.

††Understanding how to control the fire is the most important thing for the host to master (which makes using an electric coil counter-intuitive, to say the least).  At a gathering in Rikyū’s style, charcoal was added once, at the beginning, and the fire was expected to last until the service of usucha was finished (with the heat beginning to decline, so the sound of the kama began to soften, as the guests were finishing their haiken of the chaire, chashaku, and shifuku — though it should not die out completely until after they leave).

     The fundamental fires were built at dawn and dusk: 

- The dawn fire formed the basis for the morning gathering.  However, by 9:00 or 10:00 AM the dawn fire would have begun to decline, and the temperature of the kama would have dropped to sub-boiling.  So the host added charcoal to return the kama to a boil, and maintain that boil until the end of the gathering — the end of the service of usucha when done during the koicha-temae.

- The charcoal laid at dusk would have heated the kama to a boil that would have begun to decline around 7:00 PM, which was the time that guests were invited for a night gathering.  Thus here, too, the host would add charcoal to the fire at the beginning of the nigh gathering to return the kama to a boil, and keep it boiling until the gathering was ended.

- The rebuilding of the fire at mid-day was intended to maintain the kama in a near-boiling state throughout the afternoon (chanoyu, like Buddhist services, was originally not conducted between mid-day and dusk).  The so-called “shōgo-chaji" [正午茶事] that commences at noon is a modern innovation, originally suggested after Western-style eating habits were becoming common beginning in the Meiji period.  (Its popularity today is largely because it is much easier to prepare for, and clean up after, than either the morning or the evening gatherings that were the traditional times for chanoyu in the past.)

◎ Since the kama is sub-boiling when the guests arrive for the morning or night gatherings, and begins to return to a boil 15 or 20 minutes after charcoal is added to the ro, there is no time for an elaborate kaiseki such as Jōō favored when he first proposed the cha-kai. (Jōō’s large kama, which holds fifteen mizuya-bishaku of water, takes more than an hour to reach a boil if started from cold water.)  This is why Rikyū usually limited the food served to little more than hassun and sake (what exactly he served depended on the circumstances governing that particular gathering, and on occasions he did offer a more substantial meal; but in his last years, he seems to have preferred to truncate the service of food dramatically so that it would fit into the schedule that was dictated by the time it took for the kama to return to a boil).  Rikyū considered the drinking of sake an important preparatory rite to the taking of koicha, and held that even monks (who are supposed to refuse all intoxicants) should have two drinks.  Likewise, the kashi are intended to prepare the guest to appreciate the taste of the koicha, and so these were always served during the sho-za as well.  The naka-dachi was not prolonged, and in this case it is possible to finish with tsuzuki-usucha before the kama begins to fail.  Thus san-tan san-ro and the boiling of the kama controlled the length of Rikyū’s gatherings, and kept them from dragging on the way modern gatherings are often apt to do.

‡‡In the Edo period ignorance over the meaning of Rikyū’s teachings caused the Senke to change the kanji with which san-tan san-ro is written to san-tan san-ro [三炭三露], and reinterpret this expression to mean “the three times the charcoal is manipulated during a Jōō-style gathering” (where the kama is always cold when the guests arrive; now san-tan is understood to mean either “the shita-bi [下火], sho-zumi [初炭], and go-zumi [後炭]” — the latter word refers to rebuilding the fire between the koicha-temae and usucha-temae — or “sho-zumi [初炭], go-zumi [後炭], and tome-zumi [止炭]” — tome-zumi means adding some more charcoal to the fire after the service of usucha is finished, “to encourage the guests to stay on and chat for a while before leaving” — though this is contradictory to Rikyū’s teaching that the wabi small room is not the place for guests to linger and chat; once usucha is finished they should take their leave), and “the three times that water is splashed in the roji" so it is always just beginning to dry each time the guests walk through (i.e., 15 minutes before the guests are scheduled to arrive, 15 minutes before they will leave for the naka-dachi, and 15 minutes before the gathering is expected to end:  naturally this assumes that the hos has at least one assistant, since he can hardly be splashing water in the roji while he is officiating in the tearoom…though Rikyū said the host should do it all by himself).

¹⁰The mizu-kame should hold enough water to fill a large mizusashi (these hold 6 or 7 mizuya-bishaku of water), mizu-tsugi (around 4 mizuya-bishaku of water), a kama (most modern kama hold between 6 and 9 mizuya-bishaku — though some of the old ro-gama from Jōō’s period hold up to 15 mizuya-bishaku of water(), plus 4 mizuya-bishaku to fill the chakin-dari (with 4 or 5 remaining so the mizu-kame will not be depleted).  Around 2 gallons may be considered the minimum volume*, and some of the old ones hold considerably more.
*Since most modern mizuya have plumbing, and since most tea practitioners are women, the trend during the twentieth century was toward increasingly smaller utensils, and many modern-made mizu-kame are found to hold less than two gallons.  The result is that the water if often too cold in winter, and the mizusashi leaves a ring on the mat (or the tana) when it is removed.  This is one reason why lacquered tana have come to replace the unlacquered ones that were preferred in earlier days.

¹¹Especially on the outside, since the oil of the hands can stain the surface of the iron and change its color.  Often this is not apparent at the time, and becomes noticeable only much later, after the kama has been stored.  Unfortunately, it is not possible to clean the outside of the kama with anything abrasive (even the kiri-wara that is used to scour the bottom), since this, too, may change the color of the metal.  For this reason the kama should be handled only with kan, or with a towel kept between the hands and the sides of the kama.

¹²The kama-sue [釜据え] is a square frame-like object on which the kama is rested on the sunoko.  This elevates the kama above the surface of the sunoko, so it can be washed without danger of back-flow washing impurities into the kama's mouth.

     Originally there were two sizes of kama-sue, but in recent years only the larger kind is commonly sold by utensil shops.  This makes it difficult to use smaller kama (such as the small unryū-gaama), and kama with a long neck such as the tsuru-kubi kama.

¹³Gan-ro [丸炉] — which means “round ro" — is the name of the sunken fireplace that is located in the mizuya.  It is very simple, just an iron ro-dan 8-sun in diameter, with a 5-bu wide flange cast around the upper edge (to support it when lowered into the floor), and fits into a round hole cut in the floor (in much the same way as the old rice cooking-pot — from which it was probably derived — that was used in the tearoom before Jōō introduced the kabe-rodan patterned after those found peasant dwellings). 

     Its purpose is not to heat the mizuya (which originally was open to the elements on two or three sides), but used to keep other things hot — shallow pots of soup or other dishes that must be kept warm until they are served, and the empty kama (so it will dry).  Embers are also banked in the gan-ro when the tearoom is closed for the night, to be used to rekindle the fire in the ro the next dawn.  (According to ancient custom, it portents bad fortune if this fire goes out.  By tradition the fire — not only for the tearoom, but for the household itself — was brought from the local Shintō shrine, or Buddhist temple, on the last night of the year, and this fire was supposed to be preserved until the last day of the next year.)  

¹⁴The aka-sugi mizuya-bishaku is only used to dip cold water from the mizu-kame.  Hot water must be handled with a bamboo hishaku — usually the same one the host used for his temae.

¹⁵Electric heating elements are not good, since they are too hot and may cause the iron to rust.  Natural gas should never be used, since it is even more damaging to iron.  Because the embers remove the oxygen from the air as they burn, this anoxic heat helps prevent the formation of orange iron oxide, and so s the best source of heat both to boil the kama, and to dry it afterwards.

¹⁶Someone wrote to ask how rust can be removed from a kama.  The easiest way I have found, which is quite effective if the rust is not so deep that the metal has become diseased, is to fill the kama with water and then boil a package of inexpensive green tea (such as Japanese sen-cha [煎茶] and Chinese ching-cha [青茶]) for several hours — or overnight if possible.  The heating element should be set to low heat, just hot enough to boil the water gently.  This will change the orange rust [Fe₂O₃] to black/brown rust [Fe₃O₄]:  the former will damage the kama, while the latter forms a protective coating that discourages further rusting.

(Video)  Preparing usucha during the koicha temae.  This is called tsuzuki-usucha [續き薄茶], “continuing with usucha.”  The video begins after the teishu-shōban and the cleaning of the chawan with hot and cold water (rubbed with the host’s thumb), and the entire process depicted in this video¹ is repeated again and again until no more usucha is needed.  (Each guest may have up to two bowls of usucha.)

◎ Because the service of usucha is taking place during a koicha-temae, using the same container of tea, the chaire and chashaku are handled in the same way as they were when serving koicha.  This is a rule of ancient standing, and one that should be respected.
¹However the reader should note that once the chakin has been placed on the lid of the kama (as happens during the course of this video), it will continue to be placed there until the end of the temae.