History of Yuba

by William Shurtleff and Akiko Aoyagi

A Chapter from the Unpublished Manuscript, History of Soybeans and
Soyfoods, 1100 B.C. to the 1980s

©Copyright 2004 Soyfoods Center, Lafayette, Californi

For updated and greatly expanded free information on this subject,
on our website go to "Historical Bibliographies and Sourcebooks on
Soy," then click on the corresponding subject. A lengthy digital book will
appear in PDF format. It is searchable using Adobe Acrobat or Adobe

Reader .

When soymilk is gently heated in shallow, open pans at 80-90*C, a cream-yellow, bland-flavored, concentrated protein-lipid surface film (called yuba) gradually forms. The films are successively removed from the soymilk surface using a long skewer, hung to air dry, and marketed as fresh or dried sheets, sticks, or chips, or made into a variety of meatlike textured protein products.

Etymology. Yuba is a Japanese term, derived from Chinese. Prior to 1979 the Wade-Giles system transcribed the Chinese term for yuba from standard Mandarin into English as toufu p'i . After 1979 the pinyin system transcribed it as doufupi . This means "tofu skin or tofu film," namely the film that forms on the soymilk if it cools in the curding vat during the process of making tofu. It is less commonly known as doufu-i ("tofu robes") and youpi ("oil skin"). According to the Pen-chao shih-chien (1695), at that time (and perhaps in former times), this food was most widely called doufu-lao , but was also called doufupi . The character lao refers to either an old woman or a wet nurse. In China, yuba is also very widely sold in V-shaped dry rolls called fuzhu ; fu is the second character in the word tofu and zhu means "bamboo," since each side of the V resembles a bamboo shoot. We might call this form "bamboo yuba."

In the Japanese word yuba, as presently written, the character yu means "hot water" and the ba means "leaf" or "flat thing." The etymology of the present term, however, is both obscure and complex, being intricately linked with the early history of the food itself. Reference to this food first appeared in Japan in the Pen-chao shih-chien ( Honcho Shokukan in Japanese), published in 1695, as noted above. This Chinese book was quite popular in Japan. When the Japanese read the Chinese characters for yuba, doufu-lao , they pronounced them tofu no uba . As stated above, lao or uba means "old woman" or "wet nurse." The next mention of this food was in the Wakan Sanzai Zukai (1711), another book containing strong residual Chinese influence. Here it was referred to as doufupi , the present Chinese term, but interestingly the text notes that "The wrinkled look of the film resembles (the skin of) an old woman." This explanation seems to imply that the earlier term lao or uba was used because of the similarity of yuba and an old woman's face. The term yuba first appeared in 1813 in the Kyonan Rubetsu-shi . This new term's origin was explained as follows:

On Yuden Mountain, a sacred holly mountain in the feudal province of Dewa, there was an inn, Kinshiya Inn, visited by many pilgrims who came to the mountain to pray. The esteemed vegetarian cuisine was prepared by an old woman who was very skilled at making a variety of delicious foods from ground soybeans and soymilk. One of the foods was a thin film, which she fried before seasoning it. Others liked it so much that they began to make it themselves here and there. Sometimes they made it in the shape of bags and squares. In Kyoto, the capital, they used it in vegetarian dishes. Since the new food had been created by the old woman of Yuden Mountain, people named it yuba, where yu is taken from the Yu of Yuden mountain, meaning `hot water' and ba is the character for `old woman.'

Note that while the word was pronounced yuba as it is now, the second character was different.

The Kotto-shu , by SANTO Kyoden, published in 1815, offered a different etymology:

Doufupi has traditionally been called yuba, however this is a mispronunciation. Its real name is uba. Some people say that the reason this film is called yuba [written with the characters `hot water' plus `old woman'] is due to its yellowish color and the wrinkled surface, which resembled the face of an old woman. However this is not true. According to the Iseitekun Ourai (written several decades earlier) yuba is called tofu uwamono (literally `tofu upper substance') and this is the correct name, since yuba is a film which forms on the surface while making tofu. It is also abbreviated as tofu-no-ha (`tofu film'). Uba is a mispronunciation of uha, in which the sound ha is mispronounced as ba. Uba and yuba each contains the sound uba, so that it is quite understandable that the dialect yuba was created.

There are also other explanations of the origin of the term. For example, uba may be written with two characters meaning "upper" and "waves," signifying that it is the "upper waves of tofu." Or yuba may have been written with characters meaning "hot water" and "film," or "hot water" and "waves." It is not clear when or why the present characters meaning "hot water" and "leaf" came to be used, but they are thought to be of relatively recent origin. In all of these derivations there are several interesting points to be observed. First, the Japanese saw fit, for some reason, to change both the pronunciation and the characters with which they wrote the name of the product. Second, many of the explanations make it appear that yuba was developed in Japan rather than introduced from China. Third, by dropping the term "tofu" from the name, they seemed to imply that yuba was an independent food, perhaps reflecting the fact that it came to be made in special yuba shops that made no tofu.

The earliest references to this food in English were by Japanese researchers, Inouye in 1895 and Oshima in 1905. They, of course, used the Japanese name yuba , which quickly caught on. No English equivalents were attempted since there were no good concise ones, and since "yuba" was short, attractive, easy to pronounce, and easy to remember (as in Yuba City, California). The only early writer in English to refer to yuba by a Chinese name was Loomis in 1914. He referred to imported Chinese yuba as toufu p'i . Horvath 1927, writing from China, referred to it as both yu p'i and yu ba .

Early European writers generally referred to this food by a long phrase, such as "the film that forms on soymilk," etc. But gradually it has come to be called le yuba in French and das yuba in German.


China . There is, unfortunately, very little information about the early history of yuba in China. The earliest known reference to yuba in a Chinese document appears in the Pen-chao shih-chien (1695), which states, "Take the thick liquid (soymilk) and boil it in a pot. Floating on the surface in the pot will be a firm layer like wet paper. This is doufupi , commonly called toufu lao . Remove and dry in the sun to make a vegetarian food." Shinoda (1974) noted that a recipe called Toufu-p'i Paozu was served to Chinese nobility during the Ch'ing dynasty (1662-1912). It is possible that yuba was developed as much as several hundred years before its first recorded appearance, but this is only supposition.

In early times, yuba was probably heated in shallow, circular, round-bottomed pots over a bed of coals. The round sheets of yuba, each draped over a skewer, were dried slowly over a weak bed of coals. This method can still be seen in some villages in Taiwan and China.

Worldwide, yuba is most widely used in China, Taiwan, and Hong Kong. Sold fresh or dried, generally in sheets, at prices anyone can afford, it is also made into a remarkable variety of ready-to-eat foods. The "bamboo yuba" made by rolling up fresh yuba sheets, then hanging them over a line to dry, so they come out looking like a V-shaped beige rope, with each leg about 7 inches long, are also very popular, as is the thick sweet yuba that forms on the bottom of the pan during simmering and is called amayuba ??

The world's first meat analogs, made from yuba, were developed in China, probably at least several hundred of years ago, probably by Buddhist chefs in temples, monasteries, or Buddhist vegetarian restaurants. The earliest process for making these meatless meats probably consisted of rolling sheets of yuba around a filling of minced and seasoned pieces of yuba, tying closed the bundle with string, and steaming for 30-50 minutes, or until a meaty texture and flavor developed. The finished products are still widely sold as vegetarian sausages, drumsticks, liver, and the like.

In a later development, sheets or scraps of fresh or dehydrated yuba were mixed with various seasonings (usually including soy sauce) and packed into hollow two-piece molds (typically shaped like a chicken, fish, or the like), which were then clamped shut and steamed for about 50 minutes. When the form was opened . . . behold: a perfectly shaped, tender chicken--with no bones. The finished product was then usually deep-fried (whole), carved, and served. Monks would enjoy these exotic dishes on special occasions and Buddhist laymen would enjoy them on the traditional "meatless" days, the first and fifteenth day of each lunar month. During the 1970s and 1980s Shurtleff and Aoyagi (1975, 1983) reported seeing, in a number of major Taiwanese and Chinese cities, the display case of the many Buddhist Vegetarian Restaurants (as well as in the market stalls) almost perfect replicas of plucked hens, roosters, and ducks, light brown fish (complete with fins, gills, eyes, and mouth), juicy hams, tripe, liver, and rolled meats--all made from yuba and bearing such fanciful names as Buddha's Chicken or Buddha's Duck. Rich red sausage links were seen hanging in rows and deep-fried drumsticks were handsomely arranged on a large platter--together with a life-sized pig's head, made of yuba, of course. The largest selections of such "meat analogs" were found in Taipei and Beijing.

In 1921 Embry (an American) and Wang described tofu- and yuba-making in China.

The soymilk is heated for about 30 minutes. During this time, films of yuba are skimmed off. The first films are called t'ien chu (`heavenly ??'; tian zhu), those forming still later, yu p'i (`oil films'; youpi), and the last ones which form are called fu-chu (`bamboo yuba'; fuzhu), and they are the thickest and sweetest. From 15-20 such films may be removed without affecting the quality of the tofu, but as a rule the films are not removed when the manufacture of tofu is the main object.

Note that the first films removed are considered to be the best quality (the same is true in Japan) and that the last ones are used to make the V-shaped "bamboo yuba." Horvath (1926), in discussing soymilk in China observed that Chinese soymilk was usually low in fat "owing to the custom of the Chinese of removing up to 30 pelliculas rich in fat, which are sold separately at a high price." Yuba was, indeed, seen as the "butter" of Chinese soymilk.

By the 1980s there were numerous varieties of yuba. Names of several cut or folded forms are given on page ??? A number of popular meat analogs are also made from yuba, as described above. These vegetarian products are remarkably similar in appearance, texture and even taste to their animal counterparts--and they have no bones (Shurtleff and Aoyagi 1975; Wang et al. 1976). In Beijing there is a famous vegetarian delicatessen near the Peking Hotel (Quan Zu Zhai at Bamine Tsao near Wan Fujin). It shows 30 foods on the wall price list, most made from yuba or tofu, deep-fried and/or simmered in a soy sauce broth. A line of 25 people of all ages may be seen waiting to buy them, mostly for health reasons (Shurtleff 1983).

Shurtleff (1983) also reported that dried yuba is widely available in China, sold at either the "Bean Products" section of markets or in dried goods stores. The most popular type seems to be the dried rolls of "bamboo yuba" ( fuzhu ), perhaps because the combination of the original thick yuba and the rolled form prevents the pieces from breaking as easily during shipping and handling as the other delicate sheets do.

Harris et al. (1949) did a nutritional analysis of yuba ( yu pi ) and stated that it was used in soups, cooked with red meat, or used in meat stuffings. It was seldom cooked alone and was moderately expensive. He called it "soybean milk clot." A popular dish containing rolled bamboo yuba in a medium thick tan sauce is ching-sha fuzhu .

Although large, modern yuba factories started in Hong Kong and Taiwan, traditional, small-scale methods are still the only ones used in the People's Republic. Although there are no known statistics on the number of yuba shops in China, there must be thousands, perhaps tens of thousands.

In the early 1970's, a number of large, modern yuba factories came to be built in Hong Kong and Taiwan using updated methods: the soymilk was heated with steam from a boiler, the soymilk was run into the steaming trays using pipes and gravity flow, the yuba was made in both round and square compartments, and much of it was dried indoors using dry steam heat. Production costs dropped, output skyrocketed, and the yuba began to be sold worldwide, mostly in overseas Chinese food stores and restaurants.

Chinese-style yuba eventually began to spread to areas with large Chinese populations in Southeast Asia. By the late 1970s in Malaysia, for example, one could find dried yuba ( t'im chok ), vegetarian sausage ( chak tie ), vegetarian duck ( chai ak ), vegetarian salted fish ( chai kiam hu ), or vegetarian meat ( chai tu kar ) (Yu 1978; Ng 1979). Similar foods, such as yuba sausage, were found in Singapore.

Japan . There has been considerable speculation about when and how yuba was introduced to Japan. Some have said that it was brought by Chinese Buddhist monks as early as the 10th century, others that it was brought back by Japanese monks visiting China in the 13th century. It is also said that Masashige Kusunoki, a famous samurai, used it as provisions during the siege of Chihaya castle during the 14th century. Yet the earliest known written reference to the food appears in the Wakan Sanzai Zukai (1711), with the next known references being in the Isei Teikun Ourai (c. 1730), Kyonan Rubetsu-shi (1813), and Kotto-shu (Santo 1815). Japan's oldest existing yuba shop started in 1716, so it is quite likely that yuba existed on a commercial scale by the late 1600s, if not earlier. It may well have been used from time to time as a food in Buddhist temples centuries earlier. As noted at Etymology above, it was probably introduced from China, but it may also have been developed independently in Japan.

Japan's earliest center of yuba production and utilization was in Kyoto, the ancient capital, which remained the center during the 1980s. Yuba has the strongest gourmet image of all Japanese soyfoods, and Kyoto yuba soon developed strong associations with both the vegetarian cuisine of the Buddhists and with the elegant cuisine of the nobility and aristocracy. It soon became one of the indispensable delicacies in both Zen Temple Cookery ( Shojin Ryori ) and in the exquisite Tea Ceremony Cuisine ( Kaiseki Ryori ). In the Shojin Ryori Kondate-shi , published between 1818 and 1830, about half the recipes included yuba in one form or another, eloquently attesting to its popularity. A children's song (whose date or origin is probably in the early 1800s) sung in Kyoto near the base of Mt. Hie, the home of a famous complex of Buddhist temples, asks "What do the monks eat on Mt. Hie?" The response is " Yuba no tsukeyaki ," the name of a yuba preparation. During the 1970s and 1980s in Kyoto, in restaurants serving Shojin or Kaiseki cuisine, yuba might well appear in more than half the dishes in a typical six-course meal. Some Japanese restaurants, such as the beautiful Sorin-an near Kyoto, specialize in yuba cuisine. Gradually the Japanese developed many unique forms and ways of folding yuba, plus a number of ready-to-eat yuba delicacies (deep-fried chips, pouches, and rolls) that were unknown in China and which have become popular tourist items in Kyoto. Shurtleff and Aoyagi (1975) have given illustrated descriptions of each of these types.

The Kyoto yuba industry traces its origins to at least the early 1700s. The four oldest existing yuba shops started in 1716, 1791, 1804, and 1833. The owner of the oldest shop is now the ninth generation. A picture of one of the oldest and most beautiful shops, Yuba Han, appears on the cover of The Book of Tofu (Shurtleff and Aoyagi 1975). Kyoto's yuba shops have always been small, family-run operations, often connected with the family home. Of the 20 shops existing in 1981, 4 started during the Edo period (1600-1868), 5 during the Meiji period (1868-1912), 4 during the Taisho period (1912-1926), and only 6 started after 1926. Thus the industry is old and well established. In 1874 the German Ritter gave a nice 85-word description of how yuba was made in Kyoto, noting that a little wood ash was added to the soymilk to raise the pH.

The rise in popularity of yuba in the late 1800s stimulated interest in its nutritional qualities. The earliest known research on yuba nutrition was published in 1899 by Kano and Iishima of the Tokyo Army Medical College. In tests of digestion on humans, they found that about 92.6% of the protein, 95.7% of the oils, and 86.8% of the carbohydrates were digested. These results were first reported in English by Oshima in 1905. In 1895 Inouye did the first nutritional analysis of yuba.

By at least the early 1920s some yuba had come to be dyed yellow to make it appear more attractive. In 1923 Piper and Morse ran a photograph of a yuba shop and noted that the yuba was simmered in copper pans, a small quantity of auramine was added to produce a thick film and the finished yuba was dried slowly on a galvanized net over a charcoal fire. Auramine, a bright yellow ketonimine dye, was also used to make kiyuba or "yellow yuba"). Horvath (1927) reported that "Recently, an improved method of manufacturing yuba was patented in Japan, consisting in the use of an electric fan adjusted over the surface of a kettle containing soybean milk heated to a temperature not over 90*C." Some of the more famous early types of yuba in Kyoto were Daitokuji Yuba, Kenninji Yuba (both named after temples), and Toji Yuba (short trough-shaped pieces). Before 1926 Kyoto yuba was called Go-yuba ( go is an honorific prefix associated with the Imperial Household); thereafter it was called Kyo-yuba .

The number of yuba shops in Kyoto has gradually decreased from the peak of 67 in 1911. It fell to 35 in 1919, then climbed to 55 in 1929, and finally decreased slowly to 20 in the late 1970s. Number of employees ranged from 172 in 1911 to 71 in 1925, then back up to 151 in 1955, the last year for which we have data. Sales reached a peak of 183,506 yen in 1929 and were 52,433 yen in 1955 (Tanaka 1955). By the 1970s the Japanese yuba industry used less soybeans than any other Japanese soyfood industry, only several hundred metric tons a year (Watanabe 1969). And a typical shop used only 50-150 pounds of dry soybeans a day to make 400-1,200 sheets of yuba. One pound of dry soybeans yields about 0.5 pounds of yuba on a dry weight basis. Partly because it is still made on a very small scale by slow, traditional, labor-intensive methods, and partly because of its image as a gourmet food (rather than a food for the people, as it is in China), yuba in Japan in 1975 sold for about 15 times as much per pound (fresh or dry) as it did in China. Starting in the mid-1970s some modernization of the small shops (especially heating the soymilk with pressurized steam) took place, but most traditional yuba craftsmen prefer their traditional and very beautiful methods. In 1982 there were two yuba trade associations in Kyoto. One, consisting of the six oldest companies, was the Kyoto Yuba Kumiai ; the other, consisting of 10 shops in Kyoto and one in nearby Otsu was the Kyoto Yuba Seizo Hanbai Jigyo Kyodo Kumiai . The two do not cooperate much with each other.

In 1980, in addition to the 20 yuba shops in Kyoto, there were 3 in Utsonomiya and Nikko (the other main yuba center, 60 miles north of Tokyo), 1 in Otsu (just east of Kyoto), and an estimated 7 elsewhere in Japan, for a total of 31. Gross net sales of the yuba produced in Kyoto were 520 million yen, or about $2.3 million. This was estimated to be 80% of the total sales and production of all yuba in Japan.

Relatively little research on yuba production has been published by Japanese. Among the earlier of these were "On the Contribution of Lipid to the Properties of the Yuba Film" (Watanabe et al. 1975), "Formation of Yuba-like Films and their Physical Properties" (Watanabe and Okamoto 1976), and "Factors Affecting Protein Film Formation" (Okamoto 1978). Only the last of these three was in English.


Europe . The earliest known reference to yuba by a Westerner was by the Frenchman Paul Champion, who lived in China. In 1866 in describing the tofu making process in China, he noted that the cooked soymilk, prior to coagulation, was poured into a large vat and the foam removed with a scoop.

After standing for several minutes, the liquid becomes covered with a thick film, which is lifted off with a stick so as not to tear it, and is hung up to dry by affixing the stick to the wall. Sometimes a second film is formed, which is treated in the same manner. The material thus solidified at the surface of the liquid is employed in foods. It is either eaten fresh or dried, and the flavor is not disagreeable. The liquid which remains in the vat is coagulated to make tofu.

In 1869 Champion and Lhôte published the world's first nutritional analysis of yuba. Other early references to yuba were by Ritter (1874, an 85-word description of how it was made in Japan), Rein (1899), Loew (1906, he said it was a very popular food at Japanese funerals), and Bloch (1907; he called it a pellicule graiseuse or "oily film" removed while making tofu). Thereafter there was little mention of or interest in yuba in Europe. It has probably never been manufactured there. In 1981 extensive information (including recipes) on yuba appeared in German in Das Tofu Buch by Shurtleff and Aoyagi.

United States . In 1895 Inouye, in Japan, made the first passing reference to yuba in English. The earliest known reference by an American was by Langworthy in 1897; he published a nutritional analysis, which was reprinted by Abel of the USDA in 1900. In 1905 Oshima published the results of Japanese nutritional research in English. In 1914 Loomis noted that Chinese-style yuba ( toufu-p'i ) was being imported to the US "in the form of vitreous, brittle, yellowish sticks." In 1923 Piper and Morse published 1-1/2 pages of by far the best information to date in English, plus a nice photograph of sheets of yuba hanging to dry on sticks over steaming pans of soymilk, and two nutritional analyses of yuba by Oshima and by Nagao. In 1926 Horvath (mentioned earlier) wrote that in China up to 30 yuba films were removed from soymilk and sold separately at a high price before selling the low-fat milk. In 1927 Horvath gave a nutritional analysis of five types of Chinese and Japanese yuba, using data from the Tokyo Hygienic Laboratory, Embry, and Adolph. He also discussed new methods of yuba production in Japan.

The first known scientific study of yuba (in any country) was done in 1970 by Mr. L.C. Wu of the Department of Food Science at the University of Florida (Gainesville). He wrote his MS thesis on yuba: Lipid-Protein Films for Human Consumption . Wu (1972a,b, 1973) Bates and Wu (1975), both in the same department, published subsequent detailed studies of methods for increasing the yield and quality of yuba, and of the basic endothermic polymerization involved in yuba formation. In 1975 they discussed texturization, noting that "these films, when properly laminated, make good meatlike structures." Also, without using binders, desired flavors can be incorporated into the films. Their method of alkali extraction, while giving a high yield of yuba, gave a very low PER (1.26). Also in 1975 H.O. Jaynes and W.N. Chou of the Department of Food Technology and Science at the University of Tennessee (Knoxville) developed a method for making yuba from soy protein isolates using oven drying in Teflon-coated pans. In 1976 Farnum, Stanley, and Gray, in Canada, studied "Protein-Lipid Interactions in Soy Films."

The first extensive information about yuba in Japan and China, including nutritional information, methods of home and economical preparation, and many recipes, was given in The Book of Tofu (1975) and Tofu & Soymilk Production (1979) by Shurtleff and Aoyagi.

In the US yuba has long been served in a variety of dishes at most Chinese restaurants, where it is called "bean curd skin," the literal translation of the Chinese name Toufu-p'i . Dried yuba is also sold at most Chinese and Japanese grocery stores. The first company in America to make ready-to-eat yuba delicacies was The Soy Plant in Ann Arbor, Michigan. In the spring of 1980 Jura McDowell, a black member of the group, developed Yuba Rolls, using yuba made on a kitchen scale. It was cut into 3 1/2 by 5-inch rectangles, filled with seeds, sauteed vegetables, and seasonings, then rolled tightly. The next year a Tempeh Dog was developed, by wrapping soybeans, inoculated to make tempeh, in a sheet of yuba, then incubating the roll. It came out like a hot dog.

In 1981 the first yuba manufacturing company in the Western world, Soyfoods of America, began operation in Duarte, California, near Los Angeles. The owner, Mr. Ken Lee and his technical director Lawrence Wu, both Chinese Americans, built a very modern, semi-mechanized plant, drawing on Wu's research in modernizing yuba production. While their main market initially will be companies who now import yuba from East Asia (especially Hong Kong), they also hope to reach the Caucasian market. It will be interesting to see how they market their new product ( Soyfoods , winter 1982).

Does yuba have a future in the West? We think so. Although it is expensive, only small amounts can be used as wrappers and delectable fillings to give exciting new foods, such as tasty hors d'oeuvres, crisp yuba chips (more delicious than potato chips), and meat analogs. The development of less labor-intensive, more mechanized methods for making yuba would do wonders in aiding its increased use.