Bulletin No. 68. 
 
 U. S. DEPARTMENT OF AGRICULTURE, 
 
 •T* y^ OFFICE OF EXPERIMENT STATIONS. 
 
 A. C. TRUE, Director. 
 
 557 
 B6 
 
 A DESGRIPTIOIN' 
 
 SOME CHINESE yEfiETABLE FOOD 
 
 AND 
 
 THEIR NUTRITIVE AM) ECONOMIC VALUE 
 
 Br 
 
 WALTEE O. BLASDALE, 
 
 Insiructor in Ghemistry, UniversHi/ of California* 
 
 WASHINGTON: 
 
 GOVERNMENT PRINTING OFFICE. 
 
 1899. 
 
Class 
 Book 
 
 T'/ssJ 
 
Bulletin No. 68. 
 
 U. S. DEPARTMENT OF AGRICULTURE, 
 
 OFFICE OF EXPERIMENT STATIONS. 
 
 A. C. TRUE, Director. 
 
 4" V C> 
 
 A DESCRIPTION 
 
 OF 
 
 AND 
 
 THEIR NUTIUTIVE AND ECONOMIC YAIUE 
 
 BY 
 
 WALTER C. BLASDALE, 
 
 Insfructor in Chemisfri/, Unirersity of California. 
 
 WASHINGTON": 
 
 GOVERNMENT PRINTING OFFICE. 
 1899. 
 
 1^ 
 
 

 .ETTER OF TRANSMIITAL 
 
 U. S. Department of Agriculture, 
 
 Office of Experiment Stations, 
 
 Washinrjton, i). C, July 15, 1899. 
 
 Sir: I have the honor to transmit herewith a report by Walter C. 
 Blasclale, instructor in chemistry at the University of California, de- 
 scribing some Chinese vegetable food materials and their nutritive and 
 economic value. 
 
 These foods are used to a considerable extent by the Chinese popula- 
 tion in San Francisco and other cities in the United States, and most, 
 il not all, of them are staple articles of diet in China and the Orient. 
 It seems probable that some of the vegetables may become generally 
 and favorably known in the United States. 
 
 Very little information has been hitherto available concerning many 
 of these materials, and it is believed the report is a useful contribution 
 to the knowledge of the food of mankind. 
 
 Tlie report is respectfully submitted, with the recommendation that 
 it be published as bulletin No. G8 of this Office. 
 Respectfully, 
 
 A. C. True, 
 
 Director. 
 Hon. James Wilson, 
 
 Secretary of Ayriculture. 
 2 
 
-0 , 
 
 ^ 
 
 CONTENTS 
 
 Page. 
 
 Introduction 5 
 
 Experimental methods 6 
 
 Roots and tubers 8 
 
 Sagittaria, or arrowhead 8 
 
 Taro 13 
 
 Water chestnut 15 
 
 Sacred lotus 17 
 
 Lily bulbs 19 
 
 Chinese sweet potatoes 22 
 
 Yam bean 23 
 
 Cassava, or manioc 25 
 
 Green vegetables and cucurbits 27 
 
 Green vegetables 27 
 
 Cucurbits 30 
 
 Seeds and grains 32 
 
 Soy beans 32 
 
 Phaseolus 36 
 
 Dolichos 37 
 
 Various water plants 39 
 
 Chinese millet 41 
 
 Fruits, nuts, and flowers 42 
 
 Fungi and alga^ 45 
 
 Miscellaneous substances 47 
 
 3 
 
ILLUSTRATIONS. 
 
 Page. 
 
 Plate I. Plant from an imported tuber of Sagittaria sinensis 8 
 
 II. Drawiqgs of roots, seeds, etc 10 
 
 III. Drawings of starch grains 12 
 
 IV. Drawings of A'arious roots 14 
 
 V. Plant from a corm of J?/eoc7iari8 /«6c?'osa 16 
 
 VI. Upper portion of a plant of the black soy bean 32 
 
 VII. Mature plant of yellow soy bean " 34 
 
 VIII. I'lant of Phaseolus mumjo 36 
 
 4 
 
SOME CHINESE VEGETABLE FOOD MATERIALS. 
 
 INTRODUCTION. 
 
 A visit to the Chinese quarter of San Francisco or any of the larger 
 Pacific Coast cities will reveal to the eyes of a nonresident much that 
 is both strange and interesting. Most of the curious roots, green vege- 
 tables, seeds, and other articles of food making up the stock in trade 
 of the Chinese merchant would be totally unknown and unsalable out- 
 side of the narrow limits of the Chinese quarter. These articles are for 
 the most part of Asiatic origin, many of them being directly imported 
 from Canton, while others, though grown on American soil, are dis- 
 tinctly Asiatic in character. Their presence here can be accounted for 
 only on the supposition that they are of considerable importance in the 
 domestic life of the Chinese. Evidently they are the equivalent of the 
 materials that make up our own vegetable dietary and presumably 
 possess ail intrinsic value for such a purpose. The thought then natu- 
 rally arises, Might not some of these materials be turned to good account 
 in the American household? 
 
 Tradition assigns to the Chinese the highest attainment in the art of 
 producing from a given area the greatest amount of ibod material. 
 The latter result has been reached both by intensive cultivation and by 
 the utilization of a great variety of food plants, so that all classes of 
 soil and climate are made to yield their quota of food It would not be 
 unreasonable to suppose, therefore, that the little-known regions of the 
 Chinese Empire Irom which we have already obtained many useful 
 plants might yield still others of real economic value. It can not be 
 taken for granted, however, that all such materials, even though they 
 do form an important part of the Chinese dietary, would be desirable 
 introductions into our own. Of the numerous factors which must be 
 considered in determining this question the composition of the vegeta- 
 bles themselves is the most important. The amount of nutrients which 
 they contain may be readily determined by submitting them to a 
 chemical investigation. Such questions as digestibility, adaptation to 
 American tastes, means of utilization, and cultural features of the 
 plants themselves are all important elements in the i)roblem. 
 
 Powerful incentives which might induce one to look for new varieties 
 of food plants are not lacking. An increase in the number of vegeta- 
 bles which are in cultivation might admit of the utilization of a greater 
 diversity of soil and climate, or of a more profitable use of the regions 
 
 5 
 
already in cultivation, or again might be desirable from a strictly dietetic 
 standpoint. 
 
 It has long been accei)ted as true that the Chinese are largely vege- 
 tarian in their diet, and this apparently without serious detriment to 
 their physical development. This statement, however, is not based on 
 accurately compiled data, and, even if true, would be of no special sig- 
 nificance until we know more about the composition of the Chinese 
 vegetables. Apparently few analyses of Chinese food materials have 
 been published, though Kellner, Kagai, Murai, and others have 
 published a large number of analyses of Japanese food stuffs, many 
 of which are produced by plants in common use among the Chinese. 
 The incomplete character of some of these analyses and indeed of 
 many others is likely to lead to erroneous conclusions. It is espe- 
 cially desirable to discriminate between crude protein and digestible 
 nitrogenous compounds when dealing with vegetable substances, since 
 often as much as two thirds of the total nitrogen content consists of 
 amido ' compounds or other substances of little or no nutritive value. 
 
 EXPERIMENTAL METHODS. 
 
 It was through the consideration of such facts as the above that the 
 work des(;ribed in this artii^le was undertaken. It has been confined 
 to a. study of the most important of the vegetable food materials found 
 in the Chinese markets of San Francisco, though many substances of 
 animal origin found there might profitably be submitted to similar 
 investigation. 
 
 The work naturally divides itself into two distinct lines of research: 
 First, the botanical and horticultural features of the materials studied, 
 and, second, the extent to which these materials ore used, the method 
 of preparing and serving them, and the food value as shown by their 
 chemical (•omi)osition. 
 
 As a starting ])oiut for both lines of work it was necessary in each 
 case to identify the plant from which the product under investigation 
 was derived.' The identification proved in some instances a more diffi- 
 cult matter than might have been anticipated, for, though many notes 
 on tlif economic jdants of China are available, a large number of them 
 contain couilicting statements and several difierent Chinese characters 
 are often used to designate the same i)lant. In those instances in which 
 the plants themselves could be grown by the author little difficulty was 
 
 'In conlirmation of this etatement see Bohraer. Landw. Vers. Stat., 28 (1883), 
 p. 247. 
 
 -An especial acivnowledfjiiient is duo Prof. John K. Fryer, of the department of 
 oriental laugiiagos df the riiiversity of California, and to Mr. W. N. Fong, a student 
 in the same institution, for assistance in this part of the work and for other courte- 
 sies; also to Mr. Charles Ford, director of the botanic garden at Hong Kong, for 
 identification of the roots of r<tcInirhizuH axfjiilatns and for references to worlcs on 
 the botany of China; and to Prof. W. A. Setchell, of the department of botany of the 
 University of California, for the identilication of species of algie. 
 
experienced, but this was in some instances an impossibility. As an 
 assistance in tliis part of the work and also for the benefit of future 
 workers in the same field, the name and Chinese character for each 
 article were obtained when purchased, and these have been recorded in 
 the body of the rej^ort. It is to be noted, however, that the names 
 used in San Francisco are all in the Cantonese dialect, and both names 
 and characters often represent commercial designations rather than the 
 terms used in Chinese literature. In many instances the literary or 
 classical terms have been added. Often the names obtained from the 
 Chinese merchants gave but an uncertain clew to the name of the plant 
 producing the article, and frequently different names were used for the 
 same article. Aside from names and characters it was very difficult, 
 either from their actual ignorance or lack of interest in the subject, to 
 obtain authentic information from the Chinese consulted by the author- 
 ities in San Francisco regarding the source and method of use of many 
 of the substances examined. 
 
 On the chemical side of the work the chief difficulty encountered 
 was in the choice of the method of analysis to be employed. The 
 uncertainties involved in many of our present methods of food analysis 
 are too well known to need any extended mention here. In order to 
 make the results comparable with other investigations of a similar 
 nature, the methods outlined by the Association of Official Agricultural 
 Chemists ' have, for the most part, been adhered to and only such minor 
 modifications have been introduced as the nature of the work seemed 
 to demand. 
 
 The substances as purchased in the markets were weighed as soon 
 as possible, and, in the case of lleshy vegetables, the water content 
 reduced by drying in an air bath whose temperature did not exceed 
 7U° C. The residue was allowed to absorb what moisture it would, 
 was then weighed again, and the percentage of loss calculated. This 
 partially dried material was then ground either in a coflee mill or by 
 means of an agate mortar until it all passed through a sieve with 50 
 meshes to tlie linear inch and was immediately placed in a tightly 
 stoppered bottle and used for the various determinations. The residual 
 water was carefully determined both at the beginning and at the end 
 of the analysis by drying in a water bath until the samjile ceased to 
 lose weight. In many instances the time occupied in making the entire 
 analysis extended over a i)eriod of eighteen months, and some samples 
 showed differences of as much as 2 per cent in the two determinations. 
 In such cases the average of the two determinations was obtained and 
 this average factor was used in reducing the results of the analysis 
 to the figures corresponding to a water-free basis. A few exceptions 
 to this method of procedure are noted in the account of the results of 
 the individual analyses. 
 
 ' U. S. Dept. Agr., Division of Chemistry Bui. 46. 
 
8 
 
 For the determination of the crude protein the Kjeldalil method as 
 adopted by the Association of Official Agricultural Chemists was 
 employed. Albuminoids were determined by the Stutzer method, and, 
 with the exception of the difficulty encountered in liltering and washing 
 the cupric hydrate i)recipitate, this proved fairly satisfactory. This 
 difficulty was particularly mai'ked in those samples which contained 
 large amounts of starch. In such instances great care had to be exer- 
 cised to avoid overheating, as this gave a precipitate of such gelatinous 
 character that filtration became impossible. 
 
 Various attempts to determine the amido compounds directly, by 
 digestion with 5 per cent hydrochloric acid and distillation of the 
 resulting solution with barium carbonate or magnesium hydrate, either 
 failed to give the entire amount of the amids or were unsatisfactory in 
 the presence of albuminoids. The method of Sachsse was found equally 
 unsatisfactory. Hence it was necessary to report simply the difference 
 between the figures found for crude protein and those found for albumi- 
 noids, as "amids by difference." 
 
 Although the provisional methods recommended by the association 
 for determining the various members of the carbohydrate group are not 
 entirely satisfactory, they have been followed in general. The results 
 are believed to be more useful than a determination of the whole group 
 by difference. 
 
 The determination of crude fiber has been made according to the offi- 
 cial method, except that asbestos was used in both filtrations. The 
 method gave concordant results when applied to the same sample, but, 
 owing to the varying degrees of difficulty experienced in making the 
 filtrations, can not be applied to a series of different materials under 
 exactly similar conditions. In applying this method to the two species 
 of fungi analyzed nearly two days were required for both filtrations, 
 during which time the acid and alkali solutions were in contact M'ith 
 the material and must have caused lower results than would have been 
 obtained had the method been carried out under normal conditions. 
 
 A total of 53 samples was collected, 42 of which were analyzed. For 
 the sake of convenience they have been divided into a number of groups 
 in reporting and discussing the work. 
 
 ROOTS AND TUBERS. 
 SAGITTARIA, OR ARROWHEAD. 
 
 The use of the autumn tubers of the various species of Sagittaria, 
 or arrowhead, as articles of food is a common practice among several 
 different races, l^ewis and Clarke, in the account of their memorable 
 journey of 1804 across the western United States, make frequent men- 
 tion of the use of the tubers of Sagittaria lati/olia (called Wappato) by 
 the Chinook Indians of Oregon. Kalm' refers to the use of the roots 
 
 Pinkerton's Voyages, 13, p. 523. 
 
U. S. Dept of Agr., Bui. 68, Office of Expt. Stations. 
 
 Plate I, 
 
 Plant from an Imported Tuber of Sagittaria sinensis. 
 
9 
 
 of a species of Sagittaria by the Indians of eastern America. Parry' 
 notes the use of the tubers of iS. latifolia by the Chippewa Indians, and 
 states that in a raw state they contain a bitter, milky juice, but on boil- 
 ing become sweet and palatable. Finally Coville^ notes the use of the 
 tubers of S. arifolia and S. latifolia by the Klamath Indians of Oregon. 
 Eeferences to the use of 8. sagittifoUa and 8. sinensis throughout China, 
 Japan, and portions of India are not uncommon in books of travel 
 relating to these countries, and apparently forms of these species have 
 been in cultivation in the East from very early times. Henry ^ states 
 that the arrowhead {S. sagittifoUa) is cultivated iu all parts of China 
 for its edible tubers, and that there is also a wild form with numerous 
 varieties which produce smaller tubers. The wild form occurs in 
 almost every pool and paddy field. Smith ^ refers to the use of Sagit- 
 taria, but the inference from his statement is that it is not very widely 
 cultivated. 
 
 The tubers of two species of Sagittaria are found in the Chinese 
 market of San Francisco. Those of *S^. latifolia are not uncommon from 
 October to May, and are known to the Chinese as "t'sz ku" or "chu 
 ko," the former name being the classical one and the latter a provin- 
 cialism. The characters for them are ^y^ |^, but they differ from the 
 characters used in Chinese literature. These tubers are the prod- 
 uct of the native American plants which grow abundantly on the 
 "tule lands" bordering the Sacramento and San Joaquin rivers, but, 
 though more largely used, apparently are not esteemed as highly 
 as those of the Chinese species mentioned below. The tubers average 
 about G centimeters in length by 2 to 4 in diameter, and weigh about 
 14 grams each. They are often somewhat flattened laterally and are 
 terminated by an elongated sprout which generally exceeds the length 
 of tbe tuber itself. In color they are tawny white with often a pro- 
 nounced bluish tinge, the color being due solely to the very thin, smooth 
 skin. This is crossed by from two to four encircling lines which mark 
 the position of axillary buds. The interior is of about the same con- 
 sistency as a potato, yellowish in color, and farinaceous in taste. 
 
 Well-developed plants were successfully grown from both forms of 
 tubers, which were obtained at diflPerent seasons and from different 
 vendors, though none of these plants yielded mature fruit. The method 
 of formation of the tubers is somewhat i)eculiar. The elongated terminal 
 sprout rapidly develops and soon produces a terminal bud, from which 
 at first roots and then leaves are produced, thus forming an independ- 
 ent plant. This in turn sends out a new series of runners, which behave 
 in the same manner. The original tuber either decays at once or may 
 
 ' Plants of Wisconsin and Minnesota, Owen's Eeport, 1852, p. 619. 
 ^Contrib. U. S. Nat. Herb., 5 (1897), No. 2, p. 90. 
 3 Notes on tbe Economic Botany of China, p. 27. Shanghai, 1893. 
 iCoutribiitious-toward the Materia Medica and Natural History of China, p. 189. 
 Shanghai, 1871. 
 
10 
 
 form more shoots from some of its numerous axillary buds, which iu 
 turu give rise to other plants. Toward autumn all the plants which 
 have been thus established develoj) long rimners, each of which bears 
 a single terminal tuber and thus completes the life cycle without the 
 introduction of any sexual process. These features render the plant 
 remarkably prolific. 
 
 In the present investigation the author has been unable to study our 
 native Sagittaria with any degree of thoroughness, but it is certain that 
 there are constant differences between the native and imported tubers 
 and the respective plants which may be grown from them. Plate I 
 shows a plant grown from one of the imported tubers. The differences 
 in the two forms of tubers are shown in PI. II, figs. 3 and 5. 
 
 The specific limitations of the members of this genus are not well 
 defined, but there can be but little doubt that the first form of tuber 
 studied is the product of the different varieties of S. hit'ifoUu, while the 
 second form, in so far as can be determined in the absence of mature 
 akenes, seems to be the product of S. sinensis. Both of these species, 
 however, are sometimes regarded as synonymous with iS'. s<(gittifolia. 
 The leaf characters of the plants produced by the first variety vary con- 
 siderably, while those produced by the second form are fairly constant. 
 The former commonly produce leaves 1 meter in length with smooth, 
 slender, semiterete petioles, having obtusely pointed apices and diver- 
 gent lanceolate lobes; also flowers that agree in all essential respects 
 with those of *S'. lafifoiia. The plant is beautiful, and is well worthy of 
 a place among the ornamental aquatics. Indeed, it is frequently used 
 for this purpose by the Chinese, who grow it in vessels of water in the 
 same manner as they do the Chinese narcissus. 
 
 The jdants produced by the second form of tuber studied have 
 slightly smaller leaves, with distinctly five-ribbed petioles, a very 
 obtuse apex, and more pronouncedly divergent lobes. The ])lant has 
 a more stocky habit than the former si)ecies, and the tuber-bearing 
 runners are much shorter. 
 
 The statement is made by Parish' that <S\ sinensis has been introduced 
 by the Chinese into southern California, and by IJrandegee- that it is 
 comm(m along the Sacramento Kiver. Both these statements, however, 
 are disputed by Smith,^ and api)arently on sufficient grounds. 
 
 The tubers of S. sinensis appear in the markets in December or Jan- 
 uary, in some seasons in large quantities, and as far as could be learned 
 are always imported from Canton. They are symmetrical and nearly 
 spherical in form, but in size, color, consistency, and other characteristics 
 closely resemble tliose of the American species and the same Chinese 
 character and name are used for them. 
 
 'Zoe, 1 (1890), p. 122. 
 2Zoe,4 (lS93),p. 217. 
 
 ^Revision of Nortli Aiiiericau Species of Saf>ittaria and Ijophotocaipus, 1894, 
 pp. 9, 12. 
 
U. S. Dept. of Agr., Bui. 68, Office of Expt. Stations. 
 
 Plate II. 
 
 Drawings of Roots, Seeds, etc. 
 
 1, Seed aud young plant of the '-horn chestnut" n'mpa bispinosa); 2, Seed of Ginkgo biloba; 
 3, Tuber of Sagittnria sinensis; 4, Seed and young plant of Nelumbium speciosum; 5, Tuber 
 of Sagittaria latifolia, sprouted; 0. Fruit and seed of the Chinese olive iCanarium album); 
 r, Kernel of Canarium seed; 8, Root of Pueraria thunbergiana. 
 
11 
 
 The starch grains of both species vary from orbicuhir to ovate in 
 their jilane of greatest extent, are occasionally somewhat angled, and 
 seldom exceed 30 // in diameter (PI. Ill, iig. 1). The hilum is somewhat 
 eccentric and the striations are pronounced. With polarized light and a 
 selenite plate they exhibit a slight play of colors. 
 
 The composition of the two sorts of tubers is shown in Table 1. The 
 only previous analysis of this vegetable which has been found is rei)orted 
 by Kellner, and represents a Japanese sample of the tubers produced 
 by S. sagittifolin . This is included in the table for purposes of com- 
 parison. 
 
 Table 1. — Composition of arrowhead tubers. 
 
 
 
 a 
 
 S 
 o 
 u 
 
 '3 
 a 
 
 1 
 
 
 P4 
 
 J3 
 
 3? 
 
 § 
 
 a 
 1 
 
 1 
 
 S 
 O 
 
 .a 
 
 i 
 
 a 
 
 a 
 
 V 
 
 ■B 
 a 
 
 Sagittaria latifolia : 
 
 Original material 
 
 Water- free substance 
 
 Sagittaria sinensis: 
 
 Original material 
 
 Water free sii l)stance 
 
 Sagittaria sagitlifoha : b 
 
 Original material 
 
 "Water- free substance 
 
 p. ct. p. ct. 
 66. 88 4. 44 
 13.41 
 
 61.51 1 7.00 
 ! 18.18 
 
 P.ct. 
 3.98 
 12.02 
 
 4.71 
 12.24 
 
 5.76 
 17.38 
 
 P.ct. 
 0.46 
 1.39 
 
 2.29 
 5.94 
 
 1.29 
 3.88 
 
 P.ct. 
 0.76 
 2.29 
 
 .24 
 .62 
 
 .55 
 1.66 
 
 P.ct. 
 19.69 
 59.46 
 
 22.95 
 61). 89 
 
 P.ct. 
 
 a 2. 49 
 
 7.51 
 
 a2.26 
 
 5.87 
 
 p.ct. 
 
 0.32 
 
 .93 
 
 P.ct. 
 0.98 
 2.97 
 
 .72 
 1.87 
 
 1.18 
 3.56 
 
 P.ct. 
 2.04 
 6.17 
 
 1.69 
 
 4.38 
 
 P.ct. 
 
 2.71 
 8.19 
 
 3.31 
 7.26 
 
 00.86 
 
 
 7.05 
 21.26 
 
 c22. 93 
 CC9.21 
 
 1.43 
 4.31 
 
 
 a Sample contained no reducing sugar.s. 
 
 /Hieported by Kellner, quoted from Konig, Chemie di-r nieBscblichen Nahrungs- und Genus.smittel, 
 3. ed., 1, p. 705. 
 c Carbohydrates by difference. 
 
 A comparison of the dift'erent analyses shows that the figures 
 obtained by Kellner differ little from the results obtained in this labo- 
 ratory, and in no instance are the differences greater than might have 
 been looked for in tubers grown in different regions. All three analy- 
 ses show that these tubers contain considerably less water than the 
 majority of our commonly cultivated root crops, and, like most such 
 foods, only small amounts of fat, ash, and crude fiber. 
 
 The most striking feature is the high content of crude protein, 
 which, generally s[)eaking, is considerably h'gher than that of potatoes 
 and similar vegetables. Kellner's analysis gives only total nitrogenous 
 substance, but the other two show a remarkably high percentage of 
 albuminoids, the difference in the crude protein of the two analyses 
 being due almost entirely to nonalbuminoid nitrogenous matter^ so 
 that, judging by these analyses, the two forms are about equally valu- 
 able as far as albuminoid nitrogen is concerned. The superiority of 
 the arrowhead over the potato in this respect is still more striking 
 when it is remembered that about 50 per cent of the nitrogenous con- 
 tent of the latter vegetable is of a nonalbuminoid character. 
 
 Of the carbohydrate constituents, starch constitutes by far the most 
 important part, though a considerable amount of some member of the 
 cane-sugar group is present. In one analysis a small amount of some 
 
12 
 
 a 
 
 reducing sugar was found, but this may have heen due to a slight 
 inversion of some body of the cane-sugar group. Furfurol-yielding 
 compounds are also present, though only in small amounts. 
 
 Some attempt was made to determine the exact nature of the nitroge- 
 nous comi^ounds present. An aqueous extract of the dry residue of the 
 second sample in the table yielded 15.G6 per cent of nitrogenous sub- 
 stance; hydrochloric acid added to this extract precipitated 5.9G i)er cent 
 of the original substance, 97 per cent being of a nitrogenous character. 
 An extract of the freshly crushed tubers gave a precipitate with both 
 potassium ferrocyanid and with phosphomolybdic acid. On heating, this 
 extract gave a heavy coagulum at 74° C, but none above that tempera- 
 ture, though the filtrate from the coagulum still gave a precipitate with 
 both the i)otassium ferrocyanid and the i)hosphomolybdic acid. With 
 hydrochloric acid the extract gave a heavy precipitate soluble in a 
 large excess of the reagent. With cupric hydrate it gave a blue solution 
 and a heavy preci[)itate. When the filtrate from the extract which had 
 been coagulated by heat was evaporated somewhat and allowed to 
 stand, several large colorless crystals resembling in form those of 
 asparagin were obtained. These also gave the well-known reactions of 
 that compound. Hence, it would seem that nitrogenous constituents 
 of the arrowhead tubers consist of (1) a soluble casein-like albuminoid 
 comprising nearly per cent of the dry tubers, (2) a soluble albumen 
 coagulating at 74° C, (3) a small amount of an insoluble albuminoid of 
 an unknown character, and (4) varying auiounts of nonalbuminoids of 
 which asparagin is an important constituent. 
 
 The total acjueous extract of the dry tubers (the second sample in 
 the table) amounted to 29.18 i)er cent of the entire weight. Subtract- 
 ing from this figure those corresponding to the soluble nitrogenous 
 constituents, the soluble ash and the soluble carbohydrates, there 
 remains 3.28 per cent of a soluble substance unaccounted for. The 
 aqueous extract has a peculiar but not unpleasant odor. After coagu- 
 lating the albuminoids, filtering and evaporating to dryness, there 
 remained a brown sticky mass, very sweet in taste and having in a pro- 
 nounced degree the peculiar odor of the extract itself. Hence, there is 
 probably at least one unidentified substance which gives to the tubers 
 their peculiar taste. This could not be obtained in a pure condition. 
 
 The statement sometimes made that the arrowhead has acrid prop- 
 erties apparently refers only to the leafy portion of the plant, or has 
 arisen from a confusion with some other cultivated water plant. 
 
 The foregoing facts seem to warrant the opinion that the arrowhead 
 otters many advantages from a dietetic standpoint. When cooked in 
 the same manner as potatoes, it forms, in the author's opinion, a very 
 acceptable food, though it is not without a pronounced and characteristic 
 flavor. It is stated that some of the poorer non-Mongolian residents of 
 San Francisco have acquired a liking for it and use it freely. The util- 
 ization of the arrowhead has already been commented on favorably by 
 
U. S. Dept. of Agr., Bui. 68, Office of Expt. Stations. 
 
 o 
 
 
 
 o 
 
 © 
 
 o 
 
 o 
 o 
 
 o 
 
 o 
 
 ^(3 
 
 © 
 06 
 
 Plate 111. 
 
 o 
 
 4 
 
 
 f 
 
 » 
 
 
 
 "* O " 
 
 © © ( 
 
 • © 
 
 6 » 
 
 o 
 
 O -o 
 
 Q o 
 
 Q 
 
 Q Q 
 
 Q) 
 
 6 
 
 Drawings of Starch Grains Magnified about 480 Diameters. 
 
 1, Sagittaria sine7isis; 2, Nehimbhim speciosum (roots); 3, Lilium juponicum brownii; 4, Trapa 
 biapinosa; 5, Eurynle ferox; 6, Amorphophallus rivieri. 
 
13 
 
 Pailleaux and Bois.* The tubers might also be used in the mauufact- 
 ure of starch. They are employed for this purpose by the Chinese, 
 and it is said a fine quality of that article is easily prepared from themi 
 by the usual methods. The comparatively large areas of land both in 
 California and in other parts of the United States which are too wet 
 for the cultivation of other root crops would seem to offer an especial 
 inducement to considering the utilization of this genius of plants. 
 
 TARO. 
 
 From remote antiquity the taro has furnished an important item of 
 the food supply of the natives of southern India, Australia, portions of 
 Africa, and many of the islands of the Pacific, and is to-day one of the 
 plants most commonly cultivated throughout the Tropics. Botanically 
 speaking, the taro cultivated for food may be any one of a number of 
 species of the genus Colocasia (also known as Caladium), representatives 
 of the family Aroidfie. The species commonly designated as Colocasia 
 antiquorum, of which a large number of varieties are known, seems to 
 be the one most widely cultivated, though G. indlca, C. odor a, and 
 C. macrorhiza, the two former possibly only varieties of C. anUquorimi, 
 are said to be extensively used. Most of these species are also used 
 as ornamental greenhouse plants in temperate climates. 
 
 Two forms of the taro are sold in large quantities in the Chinese 
 market of San Francisco. Both are said to be imported either from 
 Canton or from the Sandwich Islands. The first, which is the smaller 
 form, is designated by the Chinese characters ^fj-, the second by 
 1^1115^. The two forms differ only in size, and produce plants which 
 do not show even varietal differences. Evidently they are the roots 
 of C. antiquorum. The smaller form consists of small ovoid or ellip- 
 soid roots about 10 centimeters in length, which weigh from 60 to 70 
 grams. The larger roots are about 24 centimeters in length, and weigh 
 from 500 to 800 grams. The latter form is considered superior to the 
 former by the Chinese and is sold at a slightly higher price. 
 
 The interior of the roots has about the consistency of a sweet potato, 
 and a microscopical examination reveals the i)resence of large amounts 
 of starch, which is present in the form of exceedingly small grains. 
 
 The roots are easily started into growth, and the Chinese market 
 furnishes a cheap source of supply of this well-known ornamental 
 plant. 
 
 Several analyses of species of Colocasia have been made by different 
 investigators, most of which, however, are somewhat incomplete. In 
 Table 2 are shown analyses of the two forms studied in this laboratory, 
 one of a small and one of a large root. For purposes of comparison, 
 an analysis by Kellner and one by Nagai and Murai are quoted : 
 
 ' Soc. d'Acclimat., 4. ser., 5 (1888), p. 1102. 
 
14 
 
 Table 2. — Composition of iaro. 
 
 
 
 p 
 ■£ 
 
 g 
 
 '3 
 n 
 
 a 
 
 is 
 
 
 .a 
 g 
 -2 
 
 02 
 
 a 
 
 P 
 m 
 
 a 
 
 CS 
 
 o 
 
 bo 
 
 a 
 a m 
 
 a 
 o 
 
 00 
 
 o 
 
 .9 
 
 a 
 s 
 
 a 
 
 Oolocasia antiquorum 
 (small root) : 
 Original material ... 
 Water-free snb- 
 
 Per ct. 
 74.20 
 
 Per ct. 
 
 1.70 
 
 6.60 
 
 1.89 
 5.51 
 2.09 
 10.81 
 1.43 
 9.73 
 
 Per ct. 
 1.67 
 
 6.46 
 
 1.62 
 4.72 
 1.39 
 7.18 
 
 Per ct. 
 0.04 
 
 .14 
 
 .27 
 
 .79 
 
 .70 
 
 3.63 
 
 Per ct. 
 0.27 
 
 1.04 
 
 .16 
 .48 
 .17 
 .91 
 .08 
 .54 
 
 Per ct. 
 17.95 
 
 69.58 
 
 25.32 
 73.82 
 6.52 
 33.70 
 10.40 
 70.26 
 
 Per ct. 
 
 1.15 
 
 4.45 
 
 1.86 
 5.43 
 
 P.ct. 
 .00 
 
 .00 
 
 .00 
 .00 
 
 P.ct. 
 
 0.98 
 
 3.78 
 
 .66 
 1.92 
 
 .70 
 3.63 
 
 p.ct. 
 1.31 
 
 5.10 
 
 1.10 
 3.21 
 
 .85 
 4.41 
 
 .90 
 6.69 
 
 P.ct. 
 2.44 
 
 9.45 
 
 Coloeasia antiquonim ' 
 (large root) : 
 Original material . . 
 Water-free sub- 
 
 67.51 
 
 3.30 
 9 63 
 
 Oolocasia antiquorum: a 
 Original material . . . 
 Water-free sub- 
 
 80.65 
 
 15.54 
 
 
 
 80.24 
 
 Oolocasia antiquorum: b 
 Original material . . . 
 Water- free sub- 
 
 85.20 
 
 
 
 0.12 
 
 .81 
 
 1 78 
 
 
 
 12.02 
 
 
 
 
 
 
 
 
 a Reported by Kellner, Landw. Vera. Stat., 30 (1884), p. 42. 
 
 h Reported by Nagai and Murai, quoted from Kiinig, Cbemie der menschlichen Nabrungs- und Ge- 
 nussmlttel, 3. ed., I, p. 704. 
 
 It is difficult to account for the remarkable difference between the 
 starch content reported by Kellner and that reported in the three other 
 analyses. Both the analyses reported from this laboratory were made 
 in dux^licate, with substantially the same results. Kellner states that 
 the roots contain large amounts of reducing- sugars. The extremely 
 minute size of the starch grains present renders the separation of 
 soluble from insoluble carbohydrates a very difficult matter. The water 
 extract always showed a slight turl)idity, even after several flltrations, 
 and it is possible that a considerable part of the substance reported as 
 cane sugar was really starch in a very finely divided condition. 
 
 The protein content shows a considerable variation in the four 
 samples. In all cases it is rather low, but even in the two samples 
 from San Francisco, which give the lowest results for this constituent, 
 the amount present is not far from the average of the results obtained 
 for potatoes and similar vegetables. The analyses also show tliat a 
 large percentage of the crude protein is of an albuminoid nature, the 
 proportion being somewhat greater than in the case of the potato. Most 
 of this was found to consist of a soluble albumin which coagulates at 
 74° C. A red coloring matter is present in the taro in considerable 
 amounts. 
 
 As a whole the taro is characterized by the high percentage of car- 
 bohydrates, of which starch is the most important representative, and 
 by the low percentage of fat and crude fiber. From a dietetic stand- 
 point it apparently offers no especial advantage over our commonly 
 cultivated vegetables, but where it is eaten it seems to be a satisfactory 
 substitute for them. It is favorably known to the Anglo-Saxon resi- 
 dents of the Tropics, who soon acquire a liking for it, and even in San 
 
U. S. Dept. of Agi., Bui. 68, Office of Expt. Stations. 
 
 Plate IV. 
 
 Drawings of various Roots. 
 
 1, Portion of the rootstock of Nelumbiuni xpeciosuin aud cross section of same; 2, Tuber of 
 Pachyrhizus angulatus; 3, The v. aX^^v chestnut (Eleochar is tuberosa); 4, A root of Blanihot, 
 probably M. utilissima; 5, A small form of taro (Colocasia antiquorum). 
 
15 
 
 Francisco there is a limited demand among the white races for the 
 roots. A root of G. antiquoruni is shown in PI. IV, fig. 5. 
 
 The area in the United States suited to the culture of taro is some- 
 what limited. A patch of it has been growing in the Garden of 
 Economic Plants of the University of California for a number of years, 
 and has produced an abundance of large sound roots. In southern 
 California it makes a rapid growth, but requires an abundant supply 
 of water. The Florida Experiment Station ^ has also experimented 
 with it and reports satisfactory results. 
 
 WATER CHESTNUT. 
 
 Though several species of Scirpus and the allied genus Eleocharis 
 bear tuber-like roots, they are not commonly considered plants of 
 economic importance by Europeans, and are seldom mentioned outside 
 of the systematic manuals. The water chestnut {E. tuberosa) is said to 
 be widely cultivated in China and Japan as a food plant, and is there 
 highly esteemed. In the former countries it is known to the English- 
 speaking residents as the water chestnut, which is but a free transla- 
 tion of " ti leh," one of its Chinese names. Smith- says that the tubers 
 are largely cultivated and sold for food all over China. "They grow 
 wild in IIui)eh in watery places, and are not often especially planted. 
 They are sweet, juicy, and resemble the cliestnut in flavor. * * * a 
 kind of arrowroot |i. e., starch] is made from them." It has also been 
 stated ' that Scirpiis fuberosus is cultivated in paddy lands for the sake 
 of its tubers, which are eaten eitlier raw or boiled. Bretschneider'* 
 states that it is cultivated all over China. 
 
 The corms of the water chestnut, for they can scarcely be called tubers, 
 begin to arrive in San Francisco markets in the latter part of May, and 
 are to be found on sale throughout the remainder of the year. They are 
 slightly flattened, about 5 centiuieters in diameter, and average about 
 13 grams in weight. They are surrounded by a rather thick cliestnut- 
 brown skin, but within are nearly white and of a somewhat watery 
 consistence. Under favorable conditions those purchased in San Fran- 
 cisco will grow readily and produce an abundance of slender terete 
 culms terminated in some instances by many-flowered spikes. Plate V 
 shows one of these plants. In taste the corms are exceedingly sweet, 
 and also i)ossess a peculiar but not unpleasant flavor. Their only 
 resemblance to the chestnut is in the color of the outer skin. To the 
 Chinese they are known as "ma hai," and are designated by the char- 
 acters ,f^]Ej^, which difiter from any of those used by Bretschneider for 
 
 • Florida Sta. Rpt. 1896, p. 9. 
 
 2 Contributions Toward the Materia Medica and Natural History of China, p. 92. 
 Shanghai, 1871. 
 ■' Useiul Plants of Japan, p. 27. Tokyo, 1895. 
 "Jour. China Branch Roy. Asiatic Soc, 25 (1890-91), p. 47. 
 
16 
 
 the plant. A drawing of three water chestnuts is shoirn in PI. IV, 
 fig. 3. 
 
 As far as observed no analysis of this vegetable has been published. 
 The composition of two samples purchased at different seasons is 
 recorded in Table 3. 
 
 Table 3. — Composition of water chestnut. 
 
 
 
 
 m 
 
 c 
 
 
 
 
 M 
 
 
 
 "? 
 
 
 1 
 
 00 
 
 a 
 o 
 
 "2 
 
 'o 
 
 a 
 
 a 
 
 to ^ 
 
 ■g=s 
 
 < 
 
 ■^ 
 
 .a 
 1 
 
 o 
 
 a 
 
 00 
 
 ^3 
 
 a> 
 
 
 
 .£3 
 
 .9 
 
 a 
 
 -a 
 
 
 > 
 
 Ph 
 
 <\ 
 
 ^ 
 
 M 
 
 
 ^ 
 
 
 <) 
 
 t5 
 
 Eleocharis tuberosa (first 
 
 
 
 
 
 
 
 
 
 
 
 
 sample) : 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 P.ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 p.ct. 
 
 P.C«. 
 
 P.ct. 
 
 Original material. . . 
 
 77.29 
 
 1.53 
 
 1.16 
 
 0.37 
 
 0.15 
 
 7.34 
 
 6.35 
 
 1.94 
 
 0.94 
 
 1.19 
 
 3.28 
 
 W ater free s u b- 
 
 
 
 
 
 
 
 
 
 
 
 
 stance 
 
 
 6.73 
 
 5.10 
 
 1.63 
 
 64 
 
 32.30 
 
 27.94 
 
 8.56 
 
 4.12 
 
 5.24 
 
 14.47 
 
 Eleocharis tuberosa (sec- 
 
 
 
 
 
 
 
 
 
 
 
 
 oud sample) : 
 
 
 
 
 
 
 
 
 
 
 
 
 Original material 
 
 77.89 
 
 1.31 
 
 1.00 
 
 .31 
 
 .27 
 
 8.09 
 
 6.02 
 
 2.60 
 
 1.22 
 
 1.18 
 
 1.42 
 
 Water- free sub- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 5.91 
 
 4.54 
 
 1.37 
 
 1.23 
 
 36.58 
 
 27.23 
 
 11.78 
 
 5.53 
 
 5.32 
 
 6.42 
 
 
 
 
 
 
 Much difficulty was experienced in preparing the corms for analysis 
 owing to the high percentage of soluble carbohydrates. This caused 
 theni to form a sticky mass, rendering it impossible to reduce the sam- 
 ple to a finely divided condition. Consequently the figures rei)orted 
 for soluble carbohydrates in the analysis of the first samjile are proba- 
 bly somewhat low, as is indicated by comparison with the analysis of 
 the second sample in which this difficulty was largely overcome. 
 
 On igniting the ash the green color, characteristic of manganese, 
 was always observed. This color was also noted in a large number of 
 the vegetable products grown in China.' 
 
 The protein content of the water chestnut is rather low, being even 
 lower than that obtained for the taro, but in respect to this con- 
 stituent, it will still compare favorably with many of the vegetables in 
 common use. Most of the protein consists of soluble albumin which 
 coagulates at 70° C. The striking feature of both analyses reported is 
 the high percentage of soluble carbohydrates. Presumably these con- 
 sist very largely of cane sugar, though several attempts to prepare this 
 in crystalline form from water extracts of bulbs were unsuccessful. 
 
 As an article of food the roots, in the author's opinion, are very pal- 
 atable. The high percentage of sugar gives them an agreeable flavor 
 even while raw, in which state they are commonly used by the Chinese. 
 
 But little information is accessible regarding the climatic conditions 
 favorable to the development of the plant. The few specimens which 
 were grown under glass seemed somewhat tender, but it is probable 
 
 'This would seem to indicate that too much reliance should not be placed on this 
 test as a means of detoctinj; adulterations in teas. 
 
U. S. Dept. of Agr,, Bui- 68, Office of Expt. Stations. 
 
 Plate V. 
 
 Plant from a Corm of Eleocharls tuberosa. 
 
17 
 
 that the moister regions of the Southern States or of the interior valleys 
 of California would prove favorable to it. Experiments on this point 
 seem desirable. 
 
 SACRED LOTUS. 
 
 This plant is better known from the pages of the early classic authors 
 and from tradition than through modern works on economic botany. 
 A plant of such surpassing beauty, whose habit of growth is so unique, 
 could not fail to excite the admiration of all who were permitted to 
 watch its development; and to the people of Egypt and many of the 
 Asiatic countries it assumed the importance of a sacred emblem which 
 was intimately associated with their religion and poetry. In addition 
 to these more aesthetic uses, the plant has from time immemorial been 
 one of economic importance and to-day occupies a not insignificant 
 place in the domestic life of several nations. 
 
 One of the most recent accounts of the plant, as viewed in its 
 economic relations, is by Jules Grisard,' from which the following facts 
 have been gleaned. 
 
 The lotus (Kelumbium) is a native of western India, Persia, Cochin 
 China, and perhaps of Australia, growing in ponds, stagnant water, 
 and rivers. It is cultivated in the basin of the Mediterranean, where 
 there is a summer temperature of about 21° C. In Cochin China it is 
 cultivated in great vases j^laced at the doors of the houses. 
 
 The uses of the lotus are numerous and varied. The stamens are 
 employed in China as an astringent remedy and also for the toilet; 
 the petioles and peduncles furnish a viscous sap employed in India as 
 a remedy for vomiting and diarrhea; the fibro- vascular bundles of the 
 petioles are made into lamp wicks and the carpophore furnishes a popu- 
 lar remedy for blood spitting. The seeds contain a white starch used 
 largely as food (see p. 39); roasted and ground they served the Egyp- 
 tians for the manufacture of a kind of bread; in China they are used 
 in soup and also as a remedy for indigestion. They are supposed to 
 have invigorative properties when used as food by convalescents. The 
 Chinese also extract from the root a starch which they say is very 
 strengthening. A decoction of the rhizomes is used as a remedy for 
 intestinal inflammation and the rhizomes themselves also become an 
 important article of food in times of famine. 
 
 Bretschueider'^ gives an interesting account of the references to the 
 various parts of the plant in the Chinese classics, and Smith ^ gives 
 many notes regarding its varied uses. 
 
 The roots or, more strictly, rhizomes of the lotus are brought to San 
 Francisco from Canton in considerable quantities and are on sale in 
 
 ' Soc.'d'Acclimat., 1896, p. 189. 
 
 2 Jour. China Branch Roy. Asiatic Soc, 25 (1890), p. 216. 
 
 ' Contributions Toward the Materia Medica and Natural History of China, p. 139. 
 Shanghai, 1871. 
 
 3485— No. 68 2 
 
18 
 
 tbe Chinese quarter throughout the late winter and early spiiiig months. 
 As there found, the yellow nodular root stocks are often 100 centimeters 
 in length, each internode measuring from 10 to 20 centimeters in length 
 by about 7 in diameter. The interior is fleshy, but firm, and of a red- 
 dish color. The roots are traversed from end to end by a series of ten 
 or more large radially-arranged tubes, with a number of smaller ones 
 between them. The appearance of one of these roots and of a cross 
 section is shown in PI. IV, fig. 1. The roots contain an abundance of 
 starch grains, which are oval, elliptical, or narrowly oblong, and exhibit 
 a beautiful play of colors with polarized light and a selenite j>late. The 
 stratifications are pronounced. 
 
 In San Francisco the roots are known as "liu ngau," though the clas- 
 sical name for the j)lant is " lien hua." The Chinese characters for them 
 are j^l^. As far as could be learned, the roots are used in San Francisco 
 for the preparation of a kind of " arrowroot" (i, e., starch), though sev- 
 eral authors state that in China the roots are also boiled and used as a 
 vegetable, or are eaten raw in somewhat the same manner as we use 
 salad plants. 
 
 It was found impossible to raise plants from any of these roots i)ur- 
 chased in San Francisco. The terminal buds in most instances had 
 been destroyed and the roots seemed to be lacking in vitality. There 
 can be no doubt, however, that the roots in question were those of 
 Nelumhinm speciosnm . 
 
 In Table 4 are tabulated two analyses of these roots — one of a sami)le 
 purchased in San Francisco, and the second, which is included for pur- 
 poses of comparison, of Japanese roots N. nucifera [sjjeciosum] reported 
 by Kellner. 
 
 Tahle 4. — Composil'ton of loins roots. 
 
 
 
 
 03 
 
 <ti. 
 
 
 
 
 bo 
 
 
 
 i 
 
 
 
 
 3 
 
 ■^-A 
 
 
 
 h 
 
 
 •^ 
 
 
 a 
 
 
 
 
 o 
 
 frS 
 
 
 
 C8 
 
 fcn 
 
 Ml . 
 
 ^ 
 
 
 B 
 
 
 t^ 
 
 n 
 
 a 
 
 ■= 2 
 
 
 ,a 
 
 3 
 
 ■n 
 
 
 
 S 
 
 
 
 5 
 S 
 
 1 
 < 
 
 1^ 
 
 < 
 
 
 1 
 
 <0 
 
 
 5 
 
 O 
 
 i 
 
 •a 
 
 Neliimbium speciosnm : 
 
 Per ct. 
 
 Perct. 
 
 Per ct. 
 
 Per ct. 
 
 P. ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 P.ct. 
 
 p. ct. 
 
 P.ct. 
 
 Original material . . . 
 
 84.26 
 
 1.57 
 
 0.91 
 
 0.66 
 
 0.19 
 
 i.n 
 
 0.33 
 
 2.18 
 
 0.70 
 
 0.76 
 
 2.24 
 
 Water-frco sub- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 9.95 
 
 5.79 
 
 4.17 
 
 1.18 
 
 48. !)9 
 
 2.12 
 
 13.88 
 
 4.83 
 
 4.81 
 
 14.28 
 
 Nelumbo nucifera : a 
 
 
 
 
 
 
 
 Original material . . . 
 
 85.8-1 
 
 1.09 
 
 .73 
 
 .36 
 
 .20 
 
 
 bll.14 
 
 
 1.02 
 
 .71 
 
 
 W ater- tree sub- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 7.75 
 
 5.19 
 
 2.56 
 
 1.41 
 
 
 J 78. 59 
 
 
 7.20 
 
 5.01 
 
 
 
 
 
 a Reported by Kellner, quoted from Konlg, Cheiuie der inenschlichen Nahrungs- und Gennssmittel, 
 3. ed., 1. p.705. 
 ftCarbobytlrate.s by difTercneo. 
 
 The two analyses are remarkably concordant. The water content is 
 considerably higher than that of the vegetables ])reviously mentioned. 
 The protein content is rather low, and of this but little more than half 
 is jn the form of albuminoids. The latter fact is not surprising when 
 
19 
 
 it is remembered that Kinosliita' found about 12 per cent of asparagin in 
 the dry roots of -N'. micifera. The most valuable ingredient of the root, 
 however, is the starch, which constitutes nearly 50 per cent of the 
 dry substance. 
 
 A mucilaginous gum manifests itself whenever the fresh roots are 
 cut, and probably this constituent gave rise to the large amount of 
 undetermined matter found. A red coloring matter, whose nature is 
 unknown, was also present in such amounts as to color deeply the water 
 in which pieces of the root had been boiled. 
 
 Though the chemical analysis shows that the rhizomes contain valu- 
 able food ingredients, their other qualities are such as are likely to 
 prevent them from forming a satisfactory food according to American 
 standards. The roots are decidedly tough and tibrous and somewhat 
 insipid in taste. Long-continued boiling lailed either to soften them or 
 bring them into such a condition that they would be easily digested. 
 
 The plant might in all probability be utilized for the production of 
 starch, but whether the product would be equal or inferior to the 
 maranta or tacca starch is not known. The form of the starch grains 
 of N. speciosnm is shown in JM. Ill, lig. 2, 
 
 The medicinal properties which have been attributed to the various 
 parts of the plant may be seriously questioned, as the Chinese materia 
 medica presents too many incongruities to permit of placing any great 
 reliance on its teachings. 
 
 ^Experiments in the cultivation of the lotus as an ornamental aquatic 
 have been in progress in various i)arts of the United States for many 
 years and have been unexpectedly successful. It is found to tolerate 
 the severe winters of the Middle East, and the short but hot summers 
 give it an ample season in which to at least perfec^t its flowers. It is 
 stated that the plant has become naturalized in a small pond near Bor- 
 dentown, N. J.- California and the Southern States aflbrd many regions 
 that are especially suited to the growth of the plant. In the great 
 interior waterway of the former, comprising the Sacramento and San 
 Joaquin rivers, it could s(;arcely fail to be successful. Strange to say, 
 this species has proved rather more robust than our own native N. 
 luteum, whose seeds and roots, it is said, were utilized as articles of 
 food by the aboriginal races of North America.' 
 
 LILV BITLBS. 
 
 Though we are accustomed to consider lilies as plants of ornamental 
 value only, the bulbs and flowers of several species have long been used 
 as articles of food. Nitobe^ gives a very interesting account of the 
 
 ' Imp. Univ. Col. Agr. [Tokyo] Bui., Vol. 2 (189.5), p. 203. 
 ' The Culture of Water Lilies and Aquatics, Peter Henderson & Co., p. 29. 
 ^Engelnumn, Trans. St. Louis Acad Sci., 2 (1860); Newberry, Food and Fiber 
 Plants of Indians, Pop. Sci. Mo., 32 (1888), p. 31. 
 4 Garden and Forest, 9 (1896), p. 12. 
 
20 
 
 species in use by the Japanese. Of these, Lilium glehni forfts the chief 
 vegetable diet of the Ainu, an aboriginal tribe now conflued to the 
 islands of Hokaido, but L. tigrinnm and L. concolor pulchellmn are the 
 two species most commonly cultivated as articles of food. Penhallow' 
 notes the use ofL. cordifolium, and in a comparatively recent work- L. 
 tif/rinum, L. auratum, aud L. elegans are included among the i^lants culti- 
 vated for their edible roots. Bretschneider'' also notes the occurrence 
 of several species of lilies with edible roots at Peking, one of which is 
 L. Ugrinum. Davy ^ notes the use of the bulbs of L. japomcum broivnii, 
 L. cordifolium, L. Ugrinum, L. concolor pulchellum, and L. glehni, and 
 the dried flowers of L. bulbifernm and HemerocaUis gramlnea as food by 
 the Chinese and Japanese. (See also p. 44.) The author also found 
 the bulbs of L. parvum, in use by the Washoe Indians of Nevada and 
 those of Jj. pardalinum in use by Indians of nortbcirn California. 
 
 From the early part of December to the latter part of August there 
 are found in the Chinese markets of San Francisco the bulbs of a spe- 
 cies of Lilium which greatly resemble those of the well-known />. aura- 
 Uim. These are sold at tlie rate of from 10 to 20 cents a pound. They 
 are all imported from Canton. The bulbs have proved to be identical 
 with the ones sold by our nurserymen under the name of 7v. hrownli, 
 and this is a[)parently the only species sold by the Chinese merchants, 
 as a large number of bulbs purchased at different times and from dif- 
 ferent dealers have invariably yielded plants corresponding to this 
 species. Unfortunately the bulV)s are often infested with mites which, 
 either primarily or secondarily, cause the death of the plant before it 
 perfects its flowers. From a collection of over 100 bulbs only 10 perfect 
 tlowers were secured. 
 
 The name L. hrownii seems to have been first i)ublishe(l by Poiteau,^ 
 but has since been regarded as synonymous with L.japonicum*' or has 
 been reduced to merely varietal significance. The latter disposition 
 has been adopted here as being the most desirable from a cultural 
 standpoint. 
 
 What seems to be the same species may also be obtained in a dry 
 form throngliout the year, and both this and the fresh bulbs are known 
 under the name of " pak hop," and designated by the Chinese charac- 
 ters "0^^. 
 
 In Table "» the analyses of lily bulbs purchased in San Francisco are 
 reported. One of them was of fresh the other of dried bulbs. Two 
 analyses of Japanese- grown bulbs are also (pioted for purposes of com- 
 parison. 
 
 ' Araer. Nat., 16 (1882), p. 119. 
 
 2 Useful Plants of Japan. Tokyo, 1895. 
 
 'Jour. China Branch Roy. Asiatic Soc, 15 (1880), p. 179. 
 
 ••Erythea, fi (1898), p.26. 
 
 '■ Rev. Hoit., 2. ser., 2 ^1843-44), 496, ([uott'd from Index Kewensis, 3 (1894), p. 81. 
 
 *• Baker, Revision of Tulipeie, 1875. 
 
21 
 
 Table 5. — ComposUion of lihj hnlbs. 
 
 
 1 
 
 d 
 'S 
 
 I 
 
 a, 
 
 "2 
 'S 
 
 CI 
 
 a 
 
 
 
 J3 
 
 be 
 a 
 
 be 
 
 IK 
 
 b£i . 
 
 a ■X 
 
 (D 
 
 o 
 
 
 13 
 (S 
 
 a 
 
 Lilium j aponicum 
 
 
 
 
 
 
 
 
 
 
 
 
 brotvnii (dried bulbs) : 
 
 Per et. 
 
 Per et. 
 
 Per ct. 
 
 Per et. 
 
 P.ct. 
 
 Per et. 
 
 Per ct. 
 
 Per et. 
 
 p.ct. 
 
 P.ct. 
 
 p.ct. 
 
 Original material 
 
 10.16 
 
 5.57 
 
 5.00 
 
 0.57 
 
 0.37 
 
 62.65 
 
 2.84 
 
 0.00 
 
 1.64 
 
 2.68 
 
 14.09 
 
 Water- free substance 
 
 
 6.20 
 
 ,5.56 
 
 .64 
 
 .41 
 
 69.73 
 
 3.16 
 
 .00 
 
 1.82 
 
 2.98 
 
 15.68 
 
 Lilium japonicum 
 
 
 
 
 
 
 
 
 
 
 
 
 brownii (fresli bulbs) : 
 
 
 
 
 
 
 
 
 
 
 
 
 Orisiiial material 
 
 66.72 
 
 2.33 
 
 1.P5 
 
 .83 
 
 .59 
 
 17.74 
 
 4.16 
 
 .00 
 
 .75 
 
 1.24 
 
 6.42 
 
 Water free substance 
 
 
 7.01 
 
 4.50 
 
 2.51 
 
 1.78 
 
 53.40 
 
 12.51 
 
 .00 
 
 2.25 
 
 3.74 
 
 19.32 
 
 Lilium tigrinum: a 
 
 
 
 
 
 
 
 -^ 
 
 
 
 
 
 
 71.46 
 
 4.51 
 15.79 
 
 
 
 .24 
 .84 
 
 
 <;21.60 
 C75.70 
 
 
 1.04 
 3.64 
 
 1.15 
 4.03 
 
 
 
 
 
 
 
 Lilivm sp. ("Yuri"):& 
 
 
 
 
 
 69.63 
 
 3.40 
 11.19 
 
 
 
 .11 
 .05 
 
 dlO.lO 
 62.83 
 
 
 
 1.42 
 4.66 
 
 1.35 
 4.44 
 
 
 Water- free substance 
 
 
 
 
 
 
 
 
 
 
 
 
 a Reported by Kellner, quoted from Konig, Chemie der menschlichen Nahrungs- und Genassmittel, 
 3. ed., I, p. 704. 
 
 ftlleported by Nagai and Murai, quoted from Konig, Chemie der menschlichen Nahrungs- und 
 Genussmittel, 3. ed.. I. p. 705. 
 
 c Carbohydrates by difference. 
 
 d In addition the authors report in the fresh substance 0.62 per cent of glucose, 2.44 per cent of 
 pectose, and 1.92 per cent of dextrin. 
 
 Wide differences are to be noted between the composition of the two 
 samples of bulbs purchased in San Francisco and the Japanese bulbs. 
 This is not surprising when it is remembered that the bulbs were of 
 different species and were grown in different countries. 
 
 Tlie protein content is much smaller in the samples of L. brownii 
 than in the others. In all the analyses the amount of protein is some- 
 what above the average amount found in the i)otato. The percentage 
 of albuminoids is noticeably greater than in the potato. Starch forms 
 by far the most important constituent of tbe bulbs and is present in 
 sufficient amount to warrant the belief that they have a high nutritive 
 value (PI. Ill, fig. 3). The analysis quoted by Nitobe is interesting as 
 giving somewhat definite information concerning the distribution of 
 the carbohydrates, though he gives no statement as to the methods 
 used in obtaining these results. Without them an accurate knowle<lge 
 of the compounds present is scarcely possible. In the samples ana- 
 lyzed in this laboratory no cane sugar was found, though the presence 
 of an abundance of mucilaginous or peciose-like substance was easily 
 recognized. 
 
 The Chinese regard lily bulbs more as a delicacy than as a standard 
 article of diet, and the customary price is considerably above that of 
 other vegetables in common use by them. A Japanese friend informed 
 the author that they are regarded by the Jai)anese as an especially 
 desirable food for invalids and convalescents. When used for this 
 purpose the bulbs are only slightly cooked and are eaten after the 
 addition of sugar. 
 
 The bulbs sold in San Francisco, as far as the author's observations 
 go, are nearly devoid of the bitter principle which is reported to occur 
 in several species of Lilium. When simply boiled, they form a pala- 
 
22 
 
 table food, and Aiuericaus might doubtless soon become ffl'customed 
 to their use. The cultural conditions favorable to the production of 
 L. brownii or of some of the other edible specnes are not ditlicult to 
 find in our own country, though it is very doubtful whether they can 
 be grown as cheaply as our other commonly cultivated vegetables. 
 One valuable feature of the bulbs is the ease with wliich they may be 
 dried, the resulting product being quite as acceptable as the fresh 
 bulbs. The value of lilies as ornamental plants under present con- 
 ditions would doubtless prevent their extended use as food in this 
 country. 
 
 CHINESE SWEET POTATOES. 
 
 Two peculiar varieties of the sweet potato are on sale in the Chinese 
 quarter of San Francisco and are noteworthy from the fact that they 
 are the only edible roots found there which are familiar in American 
 homes. One of these is a yellow variety with somewhat angular- 
 pointed roots. It was at first supiwsed to be a true yam, or Dioscorea, 
 but by growing a plant from one of the tubers it was shown that they 
 were the produ€t of Ipomcea batatas (H^). This plant produced the 
 lobed leaves characteristic of the yam-like varieties of the sweet 
 potato, but it was impossible to connect it with any of the American 
 varieties of which descriptions were accessible.' 
 
 A second red variety of /. batatas with rounded ends, which is also 
 different from our commonly (cultivated forms, is largely used by the 
 Chinese. A plant grown from one of these roots produced the cordate 
 leaves characteristic of a large number of varieties of this species. In 
 Table is shown the comi)osition of the two sorts of Chinese sweet 
 potatoes. The comi)osition of the average of a number of American 
 varieties is quoted lor purposes of comparison. 
 
 Tablk 6. — Composition of sweet potatoes. 
 
 Ipomoea batatax (roots 
 with pointed cihIs) : 
 
 Original material 
 
 Waterlrcc sub.stauca 
 Ipomoea batatan (roots j 
 with rounded ends) : 
 Original material . . 
 W ater free su b.s tauce 
 American sweet potatoes 
 (average of 95 analy- 
 ses) ; a 
 Original material .. 
 Water-free substance 
 
 
 
 
 
 
 
 
 
 
 
 -a 
 
 ""^ 
 
 
 
 
 >i<D 
 
 o 
 
 'S 
 
 S 
 
 — a 
 
 00 g 
 
 3S 
 
 es 
 
 P 
 
 
 
 ^ 
 
 Pk 
 
 < 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 73.44 
 
 0.78 
 
 0.77 
 
 0.01 
 
 
 2.95 
 
 2.90 
 
 .05 
 
 77.47 
 
 .73 
 
 .70 
 
 .03 
 
 
 3.26 
 
 3.11 
 
 .15 
 
 69.0 
 
 1.80 
 5.81 
 
 
 
 
 
 
 
 
 
 
 
 
 its 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t* 
 
 Ml , 
 
 
 
 
 
 
 
 tB 
 
 
 
 .a 
 
 aj 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 * 
 
 a 
 
 r3 
 
 
 .fi 
 
 C8 
 
 
 « 
 
 
 h 
 
 
 h 
 
 02 
 
 O 
 
 « 
 
 (J 
 
 < 
 
 Perct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Perct. 
 
 Perct. 
 
 0.25 
 
 14.65 
 
 1.71 
 
 4.07 
 
 1.02 
 
 0.85 
 
 .95 
 
 55.16 
 
 6.43 
 
 15.34 
 
 3.85 
 
 3. 20 
 
 .22 
 
 9.67 
 
 4.02 
 
 3.19 
 
 .03 
 
 1.00 
 
 .98 
 
 42.93 
 
 17.88 
 
 14.16 
 
 4.13 
 
 4.42 
 
 .70 
 
 
 'j2G 
 
 1 
 
 1.3 
 
 1.10 
 
 2.26 
 
 
 &84 
 
 19 
 
 4.19 
 
 3.55 
 
 Perct. 
 3. 22 
 12. 12 
 
 oU. S. Dept. Agr., Office of Experiment Stations Bill. 28 (rev. ed.), p. 68. 
 b Carbohydrates by difference. 
 
 'For (lescnptiou of these, Texas Sta. liul. 28, Gciirgiii Sta. Bui, RO, .•nul Louisiana 
 Sta. Hul. 13 were consiilteil. 
 
23 
 
 As shown by their comi)ORition neither of the Chinese sweet potatoes 
 possesses any advantage over the ordinary American varieties. Both 
 of these forms of sweet potato are said to be grown in California, but 
 they are probably of Chinese origin. Apparently they possess no 
 features which would render them worthy of introduction. 
 
 YAM BEAN. 
 
 The name "fan ko" is used by the Chinese of San Francisco to desig- 
 nate two ditferent tuberous roots. One of these is globular or nai)i- 
 form in shape and varies from 1 to 'J decimeters in diameter. It is cov- 
 ered with a thick, yellow, stringy bark, which readily peels otf and 
 leaves a white fleshy interior of firm consistency and sweet taste. It 
 agrees well with tlie root described by Henry' under the Chinese nan»e 
 given above, and the characters tliere used for this name {^'^j) are the 
 same as those used in San Francisco. Henry surmises that this is the 
 root of Paehi/rhizus anfjulatns (PI. IV, tig. 2), a leguminous vine which 
 is a native of Central America and is now widely diffused throughout 
 the Tropics, althougli he was unable to secure detiiiite proof of this 
 statement. As it was found imi)ossible to cause these roots to grow, 
 a letter of imjuiry was addressed by the author to Mr. Charles J\. Ford, 
 director of the Botanic Gardens at Hongkong, regarding them. In 
 repl}^ Mr. Ford stated that he had proved by actual experiment that the 
 roots described by Henry were the product of P. angulatits. 
 
 The current descriptions of this i)lant give but little information 
 regarding the character of the root It is stated by one authority^ 
 that there is a cultivated form of the yam bean properly referred to 
 Dolichos tuherosHu, but later transierred to the genus Pachyrhizus by 
 Spreugel, which diflers from P.mujulaiHs in the more rounded, nonden- 
 tate leaves, the white and not violet flowers, and the larger pods. The 
 opinion is also advanced that this form is specifically distinct, and that 
 the name 1\ tuberotius should be adopted provisionally. Forbes and 
 Hemsley^ recognize only P. angulatus from China, and this is said to 
 be in cultivation. Fawcett^ states that the root of P. tuberosus is formed 
 of a number of simple cord-like fibers, several feet in length, stretching 
 under the sarface of the ground, bearing in their conrse a succession 
 of tubers. This description is in entire accord with the roots found in 
 San Francisco. These have been called in this bulletin P. angnlaUis 
 without, however, expressing any opinion whether the form known as 
 P. tuberosus is distinct. 
 
 The composition of P. angulatus purchased in San Francisco and of 
 P. angulatus and P. tuberosus grown in British Guiana, analyzed by 
 Harrison and Jennian, is shown in the following table: 
 
 ' Notes OQ the Economic Botany of China, p. 58. Shanghai, 1893. 
 
 - Kew Misc. Bnl., 1889, pp. 17, 62, 121. 
 
 'Index Florin Sinensis, 1887, p. 194. 
 
 ■* Economic Products of .Jamaica, 1891, p. 37. 
 
24 
 
 Table 7. — Composition of yam hean roots. 
 
 
 
 a 
 
 'S 
 
 2 
 
 2 
 'S 
 a 
 
 e 
 
 S . 
 < 
 
 
 1 
 
 
 
 5 
 
 
 3 
 O 
 
 <1 
 
 5 
 
 Pachyrhiziit angulMus: 
 
 Origiual material 
 
 Water-free substance 
 
 PachyrhizKS angulatus: a 
 
 Original material 
 
 Water-free substance 
 
 Pachyrhizus tiiherosus: a 
 Original material 
 
 Per ct. 
 78.09 
 
 85.70 
 
 82.25 
 
 Per ct. 
 
 2.18 
 9.81 
 
 Per ct. 
 1.44 
 6.59 
 
 1.04 
 
 7.27 
 
 1.05 
 5.91 
 
 Per ct. 
 0.73 
 3.35 
 
 P. ct. 
 
 0.18 
 .80 
 
 .51 
 3.56 
 
 .30 
 1.69 
 
 Per ct. 
 
 8.45 
 38.58 
 
 &2.03 
 14.19 
 
 c8. 46 
 47.68 
 
 Per ct. 
 3.71 
 16.95 
 
 1.29 
 
 7.27 
 
 Per c«. 
 
 1.84 
 8.41 
 
 1.38 
 9.65 
 
 .26 
 1.46 
 
 P. C<. 
 
 1.43 
 6.53 
 
 .71 
 4.96 
 
 .66 
 
 3.72 
 
 P. c«. 
 
 0.80 
 
 3.65 
 
 1.37 
 9.58 
 
 1.84 
 10.37 
 
 P. c«. 
 3.31 
 15.14 
 
 
 
 
 
 a Keported by Harrison and Jenmau, Rpt. Agr. Work Bot. Gard. Britiah Guiana, 1891-92, pp. 70, 71. 
 
 6 In addition,"the authors report in the fresh substance 0.17 per cent resin, 4.30 per cent digestible 
 fiber, and 2.79 ])er cent gums, pectose, etc. 
 
 c In addition, the authors report in the fresti substance 0.13 per cent resin, 2.14 per cent digestible 
 fiber, and 1.62 per cent gums, pectose, etc. 
 
 The analysis of the roots shows that they contain an abundance of 
 nutritive materials. The protein content is also reasonably high, and 
 viewed in the light of the analysis alone the roots might be pronounced 
 a valuable food. 
 
 The roots also contain considerable cane sugar and a large amount 
 of starch. The starch is present in nearly spherical grains which vary 
 much in size. Compound grains or three or more individuals are not 
 uncommon. The hilum is nearly central and is not elongated as in the 
 case of most leguminous starches. No indication of stratification 
 rings could be detected, and polarized light and a selenite plate failed 
 to show a play of colors. 
 
 The varieties from British Guiana contain much more fat and ash 
 than the Chinese specimens. The method of reporting the carbohy- 
 drates, however, differs from that used by the author, and the results 
 are scarcely comparable. Judged by these analyses the Chinese-grown 
 roots are superior from a dietetic point of view to those from South 
 America. 
 
 Harrison and Jenman' note the presence of a poisonous resin in the 
 tubers and seeds of Pachyrhizus tuberosus and a smaller amount in the 
 seeds of P. angulatm. This resin was found to be a very active fish 
 poison. 
 
 Long continued boiling of these roots failed to render them even 
 reasonably tender, and left them with a sweet but otherwise insipid 
 taste. The flesh is too tough to admit of being converted into a satis- 
 factory esculent when judged by American tastes, though it has been 
 stated by many authorities that the roots are used in this manner by 
 the Chinese. Macfadyen^ speaks in the highest terms of the edible 
 qualities of the roots, and Dr. Trimen, of Ceylon,' found that both the 
 roots and the green pods formed a valuable vegetable, but the more 
 
 'Kept. Agr. Work Bot. Gard. British Guiana, 1891-92, p. 70. 
 
 2 Quoted by Fawcett. Loc. cit. 
 
 3 Kew Misc. Bnl., 1889, p. 17. 
 
25 
 
 recent report of Harrison and Jenman^ (Britisli Guiana), does not 
 praise the edible qualities of the root very highly. Their principal use 
 by the Chinese, however, is for the preparation of a starch which is 
 said to be of a superior quality. 
 
 As far as could be learned the Chinese obtain their comparatively 
 large supply of roots entirely from Canton. 
 
 The plant requires a warm climate for its development, and probably 
 there are but few localities in the United States in which it could be 
 successfully grown. 
 
 A second root, for which the same Chinese name is used, but which 
 is imported to a limited extent only, is often 1 meter in length, with 
 a diameter of only about 12 centimeters. It also has a tough, stringy 
 bark like the former species, and is still more tough and fibrous within 
 than the former roots. There is but little doubt that this is the root 
 of Pueraria thunhergiana (PI. II, fig. 8), a leguminous vine which is a 
 native of China and Japan and greatly resembles the former species. 
 Specimens of this plant have been growing at Berkeley, Cal., for 
 several years, but do not produce roots similar to those sold by the 
 Chinese. This root, designated by the characters |^:^, is so manifestly 
 unfitted for use as an esculent that the author has not tliought it 
 worthy of an analysis. No satisfactory information could be obtained 
 from the Chinese regarding their use of it. The roots are said to be 
 used in China and Japan for the production of starch, and a fiber is 
 also prepared from the stalks of the plant. As an ornamental vine the 
 plants have many valuable features. 
 
 CASSAVA, OR MANIOC. 
 
 The characters here used ^^ might be freely translated " the ginseng 
 vegetable," and is not found in the works on Chinese botany or the 
 Chinese-English dictionaries. It is used by the Chinese in San Fran- 
 cisco to designate a peculiar root which is about 5 decimeters in length 
 and 1 in diameter and often weighs as much as 500 grams. The roots 
 taper gradually at both ends and are covered with a gray bark about 
 8 millimeters in thickness. Within this is found a white sweet flesh 
 which is traversed by a series of fine fibro-vascular bundles. Though 
 it has been found impossible to grow plants from these roots, there 
 is little doubt that they are the product of either Manihot idUisshna 
 (PI. lY, fig. 4), the tapioca jilant, or M. a'ipi, the sweet cassava — prob- 
 ably of the former species. The use of the above characters, meaning 
 the ginseng plant, as a means of designating the roots evidently alludes 
 to the habit of the plant itself, as both species of manioc present cer- 
 tain superficial resemblances to the Panax ginseng although altogether 
 different botanically. The roots on sale in San Francisco are imported 
 from China, and their principal use is in the preparation of a starch to 
 
 ' Loc. cit. 
 
26 
 
 wbicb medicinal properties are attributed, altbougb probably only 
 from tlie fancied resemblance of tbe plant to tbe true ginseng. The 
 author was also informed that the roots themselves were eaten after 
 long-continued boiling. 
 
 The Division of Chemistry of this Department has published a 
 bulletin' on sweet cassava, Its culture, properties, and uses. This is 
 based iu part on the experience of Florida agriculturists. 
 
 The Florida Station has experimented for a number of years with 
 cassava, and has recently published a number of results.- The plant 
 has been very successfully grown. The bulletin describes methods of 
 culture, the use of the roots for fattening pigs and steers, and as a food 
 for horses and cows. The roots are also used for the manufacture of 
 starch. In addition, a number of receipts are given for making pud- 
 dings, fritters, and batter cakes from the freshly grated root. The 
 cassava grown in Florida does not possess poisonous i^roperties. 
 
 The composition of manioc, manioc flour, and cassava was also 
 recently reported by Bouame.' 
 
 An analysis of one of the roots purchased in San Francisco, includ- 
 ing the bark, is here rej^orted. Other analyses are quoted for purposes 
 of comparison : 
 
 Table 8 — Composition of cassava roots. 
 
 Manihot vtilissima? : 
 Original material. . . 
 Water- free Bubstauce 
 
 Manihot a'ipi :a 
 
 Original material. 
 Water-free substaure 
 
 Manihot iitilissima : b 
 
 Original material 
 
 Water-free substance 
 
 Cassava : d 
 
 Original material 
 
 Water-free substance 
 
 Perct. Perct. 
 80. 72 1. 58 
 8.21 
 
 66.69 
 
 .64 
 1.66 
 
 1.17 
 3.62 
 
 .85 
 2.55 
 
 Per ct. 
 1.43 
 
 7.40 
 
 Per ct. 
 
 0.15 
 
 .80 
 
 P.ct. 
 0.17 
 
 .17 
 .44 
 
 .40 
 1.24 
 
 Perct. 
 12.01 
 62.28 
 
 30.98 
 80.06 
 
 23.10 
 71. 39 
 
 Perct. 
 
 0.70 
 3.95 
 
 Perct. 
 1.59 
 8.25 
 
 e29.84 
 e89.65 
 
 P.ct. 
 
 0.85 
 4.43 
 
 1.50 
 4.63 
 
 1.68 
 5.04 
 
 P.ct. 
 0.63 
 3.28 
 
 .51 
 1.31 
 
 .65 
 2.01 
 
 .74 
 2.22 
 
 P.ct. 
 1.69 
 8.71 
 
 5.52 
 14.27 
 
 C5.53 
 cl7. 08 
 
 a Reported by Wiley, TJ. S. Dept. Agr., Division of Chemistry Bui. 44, p. 9. 
 
 b Reported by Payeii, Corapt. Rend. Acad. Sci. Paris, 44 (1857), p. 401. 
 
 c Sugars, gums, etc. 
 
 d Reported by Stockbridge, Florida Sla. Bui. 49, p. 39. 
 
 e Carbohydrates by difference. 
 
 It will be seen that these roots are especially rich in starch, and con- 
 tain considerable amounts of both cane sugar and reducing sugar. 
 The fat content is low and the protein content somewhat greater than 
 that of the average potato. Nearly 90 per cent of the protein is true 
 albuminoid. It is well known that hydrocyanic acid occurs iu most 
 species of manioc, especially of M. utilissima, but the temperature at 
 
 ' U. S. Dopt. Agr., Division of Chemistry Bnl. 44. 
 
 * Florida Sta. Bui. 49. 
 
 ^'Kap. An. Sta. Agron. [Maiirifcins], 1«!)7, it.57. 
 
27 
 
 whicli the above sample was dried previous to analysis should have 
 driven off the greater part of this; and even if present, it is scarcely 
 probable that it would have materially increased the figures for either 
 crude protein or albuminoids on account of the method used for these 
 determinations. 
 
 The roots sold in San Francisco evidently differ greatly in composi- 
 tion from those analyzed by Wiley,' the ash being much greater, the 
 fat much less, the protein nearly four times as great, and the starch 
 and sugars somewhat less. The analysis of roots of M. utilissima 
 quoted from Payen^ more nearly approaches that of the San Francisco 
 sample. 
 
 The starch grains found in the root purchased in San Francisco dif" 
 fer from those commonly figured in illustrating the starch of the sago^ 
 and also differ from the grains of M. a'ipi as given by Wiley.' They 
 are commonly oval or oblong and measure on the average 25 by 17.5/^. 
 The hilum is eccentric and the rings prominent. 
 
 Judged by the results of the analysis alone, if one has in view the 
 utilization of the plant as a vegetable, the Chinese-grown roots appear 
 to possess decided advantages. If they are the product of if. utilis- 
 sima, however, the presence of hydrocyanic acid in amounts sufficient 
 to render their continued use objectionable is to be feared. If the i^ro- 
 duction of starch or glucose is the o.bject desired, the roots of AF. a'ipi 
 would be the more desirable of the two species, jtidged by the analyses 
 reported above. 
 
 GREEN VEGETABLES AND CUCURBITS. 
 GREEN VEGETABLES. 
 
 The peculiar forms of cabbage and mustard used by the Chinese resi- 
 dents of our Eastern cities have already been fully described by Bailey.^ 
 Most of the forms there discussed are found on sale in San Francisco, 
 and but little need be added to what he has written regarding their 
 botanical characters. 
 
 Brassicape-Uai^ the variety of cabbage known as "pe-tsai,'" designated 
 by the Chinese characters g ^, appears to be more largely used and 
 more highly esteemed than any other. This consists of a loosely com- 
 pacted cluster of tender white leaves with greatly thickened midribs, 
 the whole resembling a head of lettuce rather than a true cabbage. 
 The central axis of the plant, however, extends to about one-half the 
 length of the head. This is the "shantung cabbage," which, though 
 long ago brought to the attention of European seedsmen and gardeners, 
 seems never to have been distributed widely enough to admit of an 
 accurate determination of its merits. The reports of Bailey and others * 
 
 ' U. S. Dept. Agr., Division of Chemistry Bui. 44. 
 ^Compt. Rend. Acad. Sci. Paris, 44 (1857), p. 401. 
 ^ New York Cornell Sta. Bui. 67, pp. 178-191. 
 * Kew Misc. Bui., 1888, p. 1.S7 ; 1893, p. 344. 
 
28 
 
 indicate that it has many valuable qualities and miglif be advanta- 
 geously brought into more general use. 
 
 Forbes and Hemsley ' refer the plant specitieally to i>. campestris; 
 Smith,^ Bretschneider, ' and apparently most writers on Chinese botany 
 refer it to B. chinensis; Loureiro/ however, regards it as a form of his 
 Sinajris peMnensis, to which he ultimately gave the varietal name of 
 "pe-tsai," and Bailey has raised this name to a specific one. 
 
 There are many varieties of "pe-tsai.'' The one sold in San Fran- 
 cisco diflers from that figured by Bailey in the more densely compacted 
 heads and in the greater spread of the leaf blades from the very base 
 of the midrib. One head weighed 750 grams. When tested, this plant 
 seemed to the author an entirely satisfactory vegetable. 
 
 The composition of " pe tsai " is shown in Table 9, p. 32. The analysis 
 there reported shows that this vegetable does not differ greatly as 
 regards food values from our commonly cultivated cabbages. The 
 water content is somewhat greater than the average of the figures for 
 American-grown samples. The protein content is also somewhat lower, 
 but nearly two-thirds of it is of an albuminoid character. The pres- 
 ence of about 30 per cent, of reducing sugar, and also of considerable 
 amounts of starch and small amounts of the cane members of the sugar 
 group, is interesting, but corresponding determinations for other forms 
 of cabbages were not found. 
 
 B. chinensis^ anothel^ largely used cruciferous plant, is sold under the 
 name of "pak-tsai," though the same Chinese character is used to des- 
 ignate it as is used for the former vegetable. The samples found in 
 the San Francisco market are made up into bundles, and consist of 
 either the young and tender leaves or the blanched petioles of the 
 leaves or the stalks derived from the central axis of the plant itself. 
 The latter sometimes show the yellow mustard-like flowers, Bailey 
 designates this plant, though with some hesitation, by the above name, 
 but Forbes and Hemsley include it under B. c^nipestris. 
 
 The comiiosition of "pak-tsai" is shown in Table 9, p. 32. In com- 
 position this vegetable shows nearly the same features as the one \}ie- 
 viously described exce{)t in the case of reducing sugars, which form 
 only about 10 per cent of the dry material. The crude protein is also 
 somewhat lower, but the figures for albuminoids are practically the 
 same. On the whole, as regards com})ositi()n the vegetable is only 
 slightly inferior to that ])reviously described, and seems Avorthy of a 
 general trial. 
 
 B. juncea. — Still a third cruciferous vegetable is found on sale 
 throughout almost the entire year in San Francisco. It consists of green 
 stalks often 50 centimeters in length, which are usually derived from 
 
 ' Index FloriB Sinensis, 1886, p. 46. 
 
 * Contributions Toward the Materia Medica and Natural History of China, p. 42. 
 Shanghai, 1871. 
 
 ^ Jour. China Branch Koy. Asiatic Soc., 15 (1880), ])p. 34, 124. 
 ••Flora Cochinesis, 1790, p. 400. 
 
29 
 
 the petioles only of B. juncea, though early in the season the entire leaf 
 is used. This vegetable is known as "kiai tsai," designated by the 
 Chinese characters ^^, and is probably a form of B. jimcea. Bret- 
 schneider' reports that several varieties of Sinajyis juncea, or B.juncea, 
 are included under the name "kiai tsai" at Peking, and gives the same 
 character as is used in San Francisco for this vegetable. No analysis 
 of "kiai tsai" was made in this laboratory, and none seems to have 
 been reported by other investigators. This species often is sold in a 
 dried form, and is also preserved in brine and used in the same way as 
 sauerkraut. 
 
 Amarantus sp. — The young seedling plants of Amarantus are in com- 
 mon use as a pot herb among thi^ Chinese of the Pacific coast. J]oth 
 our native species and a form which is said to be regularly cultivated 
 are in use. The jilants found in market are too immature to admit of 
 specific identification. This vegetable goes under the name of " in tsai," 
 designated by the characters ^^, which Bretschneider- says is gen- 
 erally applied to several species of Amarantus. Bailey^ describes a 
 form of A. f/angeticus used by the Chinese as a pot herb. Smith ^ speaks 
 of the use of ^. oleraceus and Forbes and Hemsley' state that A. blitum, 
 A. caudatus, A. {/angeticus, and A. panicidata are commonly cultivated 
 as vegetables in China. Several varieties of indigenous Amarantus 
 and the closely related Cheuopodiums are eaten to a limited extent in 
 the United States. Their value has been pointed out by Coville. •* 
 
 The composition of Amarantus purchased in San Francisco is shown 
 in Table 9, p. 32. The analysis here recorded presents no unusual 
 features save in the high content of crude protein, in which feature it 
 surpasses the cruciferous vegetables mentioned above. 
 
 Solanum melongena. — A peculiar variety of the eggplant, known in 
 San Francisco as "pak ke," designated by the characters g jjfi, is one of 
 the noticeable features of the Chinese vegetable stands during the 
 summer months. This has a perfectly smooth white skin and is from 
 to 24 centimeters long and about 10 in diameter. Osbeck, Loureiro, and 
 most of the early writers on Chinese botany speak of the cultivation of 
 the eggplant. The composition of this eggplant is shown in Table 9, 
 p. 32. Apparently but few analyses of this vegetable have been 
 made, but the results obtained from the analysis of the Chinese variety 
 in this laboratory agree quite closely with that of an American-grown 
 specimen." The percentage of sugars and starch is apparently quite 
 large, though the real nature of the carbohydrates is uncertain since 
 
 'Jour. China Branch Roy. Asiatic Soc, 15 (1880), p. 92. 
 ^Ibicl., p. 168. 
 
 3New York Cornell Sta. Bnl. 67, p. 199. 
 
 ^Contributions Toward the Materia Medica and Natural History of China, p. 12. 
 Shanghai, 1871. 
 ■'Index Floric Sinensis, 1891, p. 319. 
 «U. S. Dept. Agr., Yearbook 1895, p. 210. 
 '^U. S. Dept. Agr., Office of Experiment Stations Bui. 28, p. 38. 
 
30 
 
 they were determined by difference. Starcli, however, «s shown by 
 qualitative tests in this laboratory, is certainly present in considerable 
 amounts. 
 
 CUCURBITS. 
 
 The fruits of several species of Cucurbitaceai are extensively used 
 by the Chinese. Bailey' has already given an account of the cultural 
 features and uses of four species, all of which were found on sale in 
 San Francisco and are there used in large quantities. Further infor- 
 mation concerning these vegetables is given by Lander-* and by 
 Georgesou.'' 
 
 Momordica charantia, the "fu kwa" or "khu qua" of Bretschneider, 
 or the "la qua" of Bailey, is designated by the Chinese characters '^ 1^. 
 This vegetable is noted in nearly all works descriptive of the botany of 
 China, where it is largely cultivated and to which country it is probably 
 indigenous. It is also sometimes called "lai kua," or the lei)rosy gourd. 
 Its immature fruit is largely used throughout the Trojjics as a condi- 
 ment in the j)reparation of curries, etc. By the Chinese it is used espe- 
 cially in the preparation of salads, etc. The closely related M. haJsamina 
 is already somewhat widely known in America as an ornamental vine, 
 and the seeds of both species are to be obtained from American seeds- 
 men. The shuttle-shaped green fruits of M. clmraniia are about 2 deci- 
 meters in length, covered with rows of wart-like tubercles, and are borne 
 on wiry stems of about the same length. Like all the other green 
 vegetables described above, they are grown in large quantities in the 
 Chinese gardens along the Sacramento River. 
 
 The M. eharanfia can scarcely be regarded as a food i>lant in the nar- 
 rower sense of that term, as the i)rincipal value of its fruits depends 
 upon the flavor they impart to other preparations, rather than on the 
 amount of nutriment they contain. Nevertheless, large amounts of the 
 fruits are eaten, and they must be considered in studying the dietary 
 of the Chinese. The analysis here reported (Table 9, p. 32) shows that 
 the fruit is not lacking in nutritive constituents, though the exact nature 
 of the compounds represented could not be determined without a more 
 extended investigation. The crude fat in this case apparently consisted 
 almost wholly of chlorophyll. Of the 18 per cent of nitrogenous con- 
 stituents, about two-thirds was of an albuminoid nature. Keducing 
 sugars, or at least soluble carbohydrates having reducing ])0wer, were 
 found to be present in con sidei able quantities, and true starch was also 
 present in relatively large amounts. 
 
 Luffa acxtangula, is known as "sz kwa," and both this and the fol- 
 lowing species are designated by the Chinese characters ^^^J^. This is 
 a native of tropical Asia, though naturalized in certain regions of 
 America. The name above given is synonymous with L. fcetida and 
 
 I New York Cornell Sta. Bui. 67, pp. 191-196. 
 •''Garden and Forest, 1 (1888), p. 483. 
 :' Amer. Garden., 13 (1892), p. 526. 
 
31 
 
 with Cucuniis acutangulus. The plant produces a green, pronouncedly 
 ten-ribbed fruit, about 3 decimeters in length, which is obtusely termi- 
 nated at the farther end, but tapers gradually to the point of union with 
 the peduncle. In uses and culinary qualities it much resembles the 
 common cucumber. In San Francisco, it is said, the Chinese use it 
 for thickening soups. Bailey' states that it is eaten raw, and also 
 cooked in the same manner as we use squashes. As shown by analysis 
 (Table 9, p. 32), it does not difter greatly from the species previously 
 described, the chief difference being in the lower percentage of nitrog- 
 enous substances and in the greater content of carbohydrates, espe- 
 cially of reducing sugars. 
 
 Liiffa cylindrica is called by the Chinese "po kua." This is the "sua 
 kwa" of Bailey.- This plant is also widely cultivated throughout the 
 Tropics, but its native country is unknown. It is synonymous with 
 L. (vgyptica, L. petiola, and Momordica cylindrica. It is reported from 
 China by most of the naturalists who visited that country in early days, 
 as well as by thB more recent botanists. Its slender, (;ylindrical, 
 crooked, yellowish-green fruits, which are often decimeters long, are 
 used in the immature condition in the same manner as we use squashes. 
 The interior of the mature fruit is filled with a fibrous mass, which 
 when dried forms a useful household article. Hence the plant is often 
 kno\\n as the ''dish-cloth" or " towel gourd," and is more often grown 
 for its fibrous interior than as a food. In San Francisco, however, it is 
 largely used for the hitter i^urpose, being prepared in much the same 
 manner as squashes are treated in American households. As shown 
 by analysis (Table 9, p. 32), it is on the whole about equivalent in food 
 value to the previously-described species, the protein content being 
 somewhat less, while the percentage of reducing sugars and starch is 
 somewhat greater. As shown by comparison, it is rather inferior to 
 the common varieties of squash, and it can scarcely be regarded as 
 more than a passable substitute for them. 
 
 Benincasa cerifera. — This plant is known as "zit kwa," and is desig- 
 nated by the characters fp JJS^. It is the same as the "tung kua" of 
 Bretschneider,' designated by the characters :^jri5;^. Under the name 
 of "Chinese preserving melon," it is already somewhat known in the 
 United States, It is a tropical vine, long cultivated in China, Japan, 
 India, and Africa, where it is often met with growing spontaneously, 
 though its native country is unknown. The young green hairy fruits 
 when about 20 centimeters in length, are used in the same manner as 
 we use the squash. The mature fruits often attain a weight of 12 
 kilograms, are perfectly smooth, and covered with a white wax. With 
 the exception of a small amount of pulpy substance which is filled with 
 the white seeds, the interior is made up of a white solid fiesh. The 
 
 > New York Cornell Sta. Bill. 67, p. 196. 
 
 2 Ibid., p. 195. 
 
 3 Jour. China Branch Roy. Asiatic Soc, 15 (1890-91), p. 153, 
 
32 
 
 mature fruit is used by the Chinese in the preparation opf confections 
 and is also said to be used as a vegetable. The analysis reported 
 below presents no especially interesting features save the high per- 
 centage of reducing sugars. 
 
 The composition of the green vegetables and cucurbits discussed 
 above is shown in the following table: 
 
 Table 9. — Composition of green vegetables and cucurbits. 
 
 GREEN VEfiETABLES. 
 
 Brassiea pe-Uai: Per ct. 
 
 Original material 95.74 
 
 Water-l'ree substance 
 
 Brassica chinciisis: 
 
 Original material 96. 55 
 
 Water-free substance 
 
 Amarantus sp. : 
 
 Original material 91.52 
 
 Water-free substance 
 
 Solanum ■)nelon<)ena: 
 
 Original material... 89.62 
 Water- free substance 
 
 CDCUEBITS. 
 
 Momordica charantia: 
 
 Original material 93.61 
 
 Water-free substance 
 
 Liiffa cylindrica: 
 
 Original material 94.66 
 
 Water- freesubstance 
 
 Luffa acutanguln: j 
 
 Original material 94.90 
 
 Water-free 8iib.stance , 
 
 Benincasa cerifera: j 
 
 Original material [ 96. 24 
 
 Water free substance 
 
 Perct. 
 
 1.19 
 
 28.07 
 
 .78 
 21.96 
 
 2.61 
 30.80 
 
 1.38 
 13.25 
 
 1.18 
 18.48 
 
 .51 
 9.57 
 
 13.39 
 
 ..50 
 13.27 
 
 Per ct. 
 0.48 
 11.36 
 
 .41 
 11.43 
 
 1.67 
 19.68 
 
 1.08 
 10.42 
 
 .79 
 12.31 
 
 .54 
 10.68 
 
 Per ct. 
 0.71 
 16. 71 
 
 .37 
 10.53 
 
 .94 
 11.12 
 
 .29 
 2.83 
 
 .39 
 6.18 
 
 .13 
 2.50 
 
 .14 
 2.71 
 
 
 -a 
 1 
 
 C3 
 3 
 
 a 
 o 
 
 te . 
 .S£ 
 3 
 
 1 
 
 D 
 3 
 
 O 
 
 < 
 
 P.ct. 
 
 Per ct. 
 
 V. ct. 
 
 Per ct. 
 
 Per ct. 
 
 P.ct. 
 
 0.15 
 
 0.31 
 
 0.09 
 
 1.29 
 
 0.52 
 
 0.56 
 
 3.57 
 
 7.19 
 
 2.11 
 
 30.25 
 
 12.16 
 
 13. 28 
 
 .10 
 
 .31 
 
 .09 
 
 ■ .37 
 
 .46 
 
 .65 
 
 2.82 
 
 8.61 
 
 2.45 
 
 10.45 
 
 12.86 
 
 18.33 
 
 .36 
 
 .50 
 
 5.82 
 
 1.57 
 
 
 
 .92 
 10.81 
 
 1.54 
 
 1.56 
 18.31 
 
 .69 
 
 4.26 
 
 
 
 .30 
 
 .63 
 
 1.31 
 
 2.91 
 
 15.11 
 
 6.10 
 
 12.66 
 
 14.88 
 
 6.69 
 
 .yo 
 
 .67 
 
 .06 
 
 .60 
 
 1.07 
 
 .34 
 
 3.19 
 
 10.56 
 
 .74 
 
 9.36 
 
 16.72 
 
 5.25 
 
 .10 
 
 1.04 
 
 .12 
 
 2.15 
 
 .46 
 
 .41 
 
 3.72 
 
 19.52 
 
 2.18 
 
 40.29 
 
 8.58 
 
 7.65 
 
 .24 
 
 .36 
 
 .10 
 
 1.57 
 
 .72 
 
 .43 
 
 4.70 
 
 7.03 
 
 1.95 
 
 30.86 
 
 14.03 
 
 8.43 
 
 .16 
 
 .31 
 
 .07 
 
 .90 
 
 .57 
 
 .35 
 
 4.34 
 
 8.29 
 
 1.74 
 
 24.19 
 
 15.19 
 
 9.48 
 
 P.ct. 
 0.14 
 
 3.38 
 
 .80 
 22.52 
 
 2.55 
 30.00 
 
 2.95 
 
 28.39 
 
 2.28 
 35.69 
 
 .45 
 8.40 
 
 1.00 
 19.61 
 
 23.49 
 
 SEEDS AND GRAINS. 
 
 SOY BEANS. 
 
 Leguminous seeds and certain preparations made from them have 
 always formed an important part of the largely vegetarian diet of the 
 Chinese and Japanese. Of the legumes the soy bean, Glycine Mspida 
 {iSoja hifipida), is the most important. The soy bean has long been 
 known in Europe. Kaempfer' was perhaps one of the first i'hiropeans 
 to describe it. This plant has been cultivated many years in Europe, 
 and is coming to be quite extensively grown in the United States, 
 largely for use as a forage plant. The soy bean has been treated of in 
 a previous publication- of this Department. A large number of vari- 
 eties of the soy bean are in cultivation in China and Japan, but only 
 two were found in the Chinese markets in San Francisco, a yellow 
 and a black variety. Aside from a difference in color, the two forms 
 
 1 Amoinitatum Exoticarum, p. 837. Lemgovise, H. W. Meyer, 1712. 
 * U. S. Dept. Agr,, Farmers' BuUetiu 58. 
 
U. S. Dept. of Agr., Bui. 68, Office of Expt. Stations. 
 
 Plate VI. 
 
 Upper Portion of a Plant of the Black Soy Bean. 
 
33 
 
 apparently do not differ materially from eacli other. Tlie yellow vari- 
 ety is known as "woiig tau," and is designated by the characters ^j^, 
 while the black is known as ^'hak tau," and is designated by the charac- 
 ters Mj^. The soy bean resembles a pea rather than a bean, although 
 the botanical characteristics of the plant indicate that it is very differ- 
 ent from any of our cultivated peas or beans. 
 
 Both varieties obtained from the Chinese market in San Francisco 
 grew readily in Berkelej^, attaining a height of about 3 feet, and in 
 spite of a very dry season produced an abundant crop of seeds. The 
 appearance of two of these plants at different stages of growth is 
 shown in Pis. VI and VII. The composition of the seeds of the two 
 varieties is shown in Ta,ble 10, the average composition of American- 
 grown soy beans being quoted also for purposes of comparison. 
 
 Taule 10. — CompoMion uf soy beans. 
 
 
 
 
 
 Hi 
 
 
 
 
 in 
 
 
 
 -d 
 
 
 
 
 n3 
 
 -s . 
 
 
 
 C 
 
 s 
 
 ^ 
 
 
 a 
 
 
 
 
 C 
 
 sr. 
 
 
 
 & 
 
 M . 
 
 
 
 g 
 
 
 
 n 
 
 
 'm ^ 
 
 
 
 s 
 
 ■S £ 
 
 « 
 
 
 
 s 
 
 
 
 
 
 .d 
 
 Sf 
 
 §^ 
 
 
 
 a 
 
 
 
 o 
 
 
 8=^ 
 < 
 
 4i> 
 
 
 a 
 
 •a 
 
 3 
 
 
 
 
 ^ 
 
 ^ 
 
 < 
 
 pq 
 
 M 
 
 5 
 
 (S 
 
 n 
 
 o 
 
 < 
 
 a 
 
 Glycine hispida (black) : 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 P.ct. 
 
 p. ct. 
 
 P.ct. 
 
 P ct. 
 
 Original material 
 
 8.25 
 
 36.35 
 
 34.63 
 
 1.72 
 
 17.22 
 
 6.80 
 
 7.38 
 
 0.00 
 
 5.25 
 
 4.77 
 
 13.98 
 
 Water- free substance 
 
 
 39. 62 
 
 37.74 
 
 1.82 
 
 18.77 
 
 7.41 
 
 8.04 
 
 0.00 
 
 5.72 
 
 5.20 
 
 15.25 
 
 Glycine hispida, (yellow): 
 
 
 
 
 
 
 
 
 
 
 
 
 Original material 
 
 8.33 
 
 36.00 
 
 35.54 
 
 .46 
 
 17.87 
 
 5.87 
 
 6.55 
 
 0.00 
 
 5.43 
 
 4.75 
 
 15.20 
 
 VV atert ree substance 
 Glycine hispida (average 
 
 
 39. 27 
 
 38.77 
 
 .50 
 
 19.49 
 
 6.40 
 
 7.14 
 
 0.00 
 
 5.92 
 
 5.18 
 
 16.59 
 
 
 
 
 
 
 " 
 
 
 
 
 
 ot 8 analyses) -.a 
 
 
 
 
 
 
 
 
 
 
 
 
 10.80 
 
 33.98 
 38.1 
 
 
 
 16.85 
 19.00 
 
 h 28. 89 
 
 
 4.79 
 5.4 
 
 4.69 
 5.3 
 
 
 
 
 
 
 
 
 
 
 
 
 a U. S. Dept. Agr., Office of Experiment Stations Bui. 11, p. 120. 
 h Carbohydrates by ditt'erence. 
 
 The analyses of these two Chinese varieties of soy beans, as well as 
 of others which have been published, show that they contain an unusu- 
 ally large amount of valuable food constituents, especially protein and 
 fat, the former constituting about 39 and the latter about 19 per cent 
 of the total weight. As shown by the analysis made by the author, 
 nearly all the protein is made up of albuminoids. The figures obtained 
 by the author for cane sugar and starch indicate that a considerable 
 amount of carbohydrates was present, although starch could not be 
 detected by the ordinary test. No special attempt was made to sepa- 
 rate the cane sugar. 
 
 There has been considerable difference of opinion among investiga- 
 tors regarding the presence of starch in soy beans. The weight of 
 evidence, however, seems to indicate that, under certain conditions, 
 starch in considerable amounts may be present. In 1880 Meissl and 
 Bocker investigated soy beans and reported the presence of starch. 
 The amount of starch found was said to be less than 3 per cent. The 
 
 1 Sitzber. Math. Naturw. CI. Akad. Wisa. (Vienna), 87 (1883), pt. 1, pp. 372-391. 
 3485— No. 68 3 
 
34 
 
 starch grains are described as being extremely small and as flifferingiu 
 form from the typical bean or pea starch. About the same time Han- 
 ausek ' also reported starch in soy beans. It was found deposited 
 where the surface of the cotyledons met, and according to the author, 
 could not be detected by the usual chemical methods. The starch 
 grains were embedded in fat and gave no color reaction after prolonged 
 treatment with iodin. A more extended study of soy beans was 
 reported by Harz.- According to this author, when the beans do not 
 ripen thoroughlj , or when allowed to ripen after the vines are cut, they 
 may contain starch, some varieties being more likely to contain it than 
 others. luoyue ^ also was unable to find starch in the mature seeds. 
 
 The carbohydrates other than starch in soy beans have been studied 
 by many investigators. Stingl and Morawski ^ report the presence of 
 small quantities of dextrin in soy beans, and about 2 per cent of a 
 mixture of different sugars which could be readily fermented. Their 
 investigation was principally concerned with the diastatic ferment 
 found in the beans. A number of tests on the quantity of the ferment 
 are reported. La Yellois'' reported the presence of 9 to 11 j^er cent' of 
 material soluble in alcohol, which did not reduce Fehling's solution. 
 Tollens^ calls this substance galactan. Schultz and Frankfurt' proved 
 the presence of cane sugar in soy beans. Maxwell ° identified paraga- 
 lactan, an insoluble carbohydrate. A number of other investigations 
 on the carbohydrates of soy beans have been reported. 
 
 In 1880 Meissl and Bocker^ reported an extended investigation of 
 the proteids of soy beans. According to these investigators the beans 
 contained the following proteids: A so-called "casein" (27.G per cent), 
 albumen (0.5 per cent), a proteid precipitated by cupric oxid and potas- 
 sium hydroxid (2.5 per cent), and, in addition, a very small amount of 
 nonalbuminoid nitrogenous substance. Osborne and Campbell'*' have 
 recently reported an extended study of the proteids of soy beans. The 
 principal proteid found is called "glycinin." It is a globulin. Accord- 
 ing to these authors the beans also contain about 1.5 per cent of an 
 albumin-like proteid, legumelin. A small amount of proteose was also 
 found. 
 
 Meissl and Booker^ studied the fat of soy beans, which is said to con- 
 tain no free fatty acid and to consist almost entirely of neutral trigly- 
 
 ' Ztschr. AWg. Oester, Apoth. Ver., 22 (1884), p. 474. 
 
 "Ztschr, Allg. Oester. Apoth. Ver., 23 (1885), p. 40; Landwirtschaftliclio Samen- 
 kimde, p. 692. Berlin, P. Parey, 1885. 
 
 'Imp. Univ. Col. Agr. [Tokyo] Bui., Vol. 2, No. 4. 
 
 ■"Monatsh. Cliem., 7 (1886), pp. 176-190. 
 
 '^Compt. Rend. Acad. Sci. Paris, 93 (1881), p. 281. 
 
 '■ Handbuch dor Kohlenhydraten, vol. 1, p. 208. Breslau, E. Trewoiidt, 1888. 
 
 ■ Bar. Deut. Chem. Geseli., 27 (1894), p. 62. 
 
 » Amcr. Chem. Jour., 12 (1890), p. 51-60. 
 
 'I.oc. cit. 
 
 "Conn. State Sta. Rpt. 1897, pt. 4, pp. .374-382. 
 
U. S. Dept. of Agr., Bui. 68, Office of Expt. Stations. 
 
 Plate VII. 
 
 Mature Plant of Yellow Soy Bean. 
 
35 
 
 cerids. At a low temperature or on standing a long time the palmatin 
 and stearin triglycerids were precipitated in crystalline form. The fat 
 has a characteristic leguminous taste. When kept for two years it 
 became thick, but was only slightly rancid. Its specific gravity was 
 0.89 at 150 C. One gram of soy-bean fat required 191.S milligrams 
 potassium hydroxid for saponification. The fat has also been examined 
 by Stingl and Morawski' and by Eoelofsen^ with reference to its iodin 
 absorption. The figures obtained for the latter constant were 121.3 
 and 138.8, respectively, and these and other facts indicate that the oil 
 has poor drying properties. 
 
 Pellet ' studied the ash of soy beans in considerable detail. 
 
 Soy beans are eaten to some extent when cooked in the same ways 
 as other beans. The Chinese express an oil from them which is a 
 standard article of commerce and is largely used in cooking. The 
 principal use which they make of the soy beans, however, is in the 
 preparation of a vegetable cheese, a kind of thick sauce, and other prod- 
 ucts. An account of the manufacture of miso and soy-bean sauce in the 
 early part of the eighteenth century is given by Kaempfer.^ According 
 toPrinsen-Geerligs,^ " tao hu," or bean cheese, is jirepared froin the seeds 
 of the white variety of soy bean. These are allowed to soak for three 
 hours in water, are then reduced to a thick paste, and the mass cooked. 
 The cooked mass is strained through a coarse cloth. The filtrate con- 
 sists of a milky- white liquid containing protein and fat. As soon as 
 this becomes cool some material is added (for instance, crude salt con- 
 taining magnesium chlorid), which precipitates the proteid material, the 
 fat being inclosed in the coagulated mass. The coagulated material is 
 pressed and kneaded into small cakes. The cakes may be dipped for 
 a few moments into a saline solution of curcuma. Variations in the 
 process give rise to a number of varieties of bean cheese. This is 
 essentially the method used by the Chinese of San Francisco in the 
 preparation of the bean cheese used by them. It is sold either in the 
 form of the freshly precipitated curd or in the form of small square 
 cakes obtained by comj)ressing the former material. It is usually 
 cooked in peanut oil before being eaten and, in the author's opinion, is 
 a palatable food. A partial analysis of one of these cakes gave water 
 81.35 per cent, fat 5.19 per cent, and ash 0.80 per cent. 
 
 The filtrate from the cooked soy beans resembles milk, and, on heat- 
 ing, a skin, not unlike that formed on milk, rises to the surface of it. 
 
 A thick sauce called "tao yu," resembling the Japanese "shoju," is 
 also prei)ared from the soy beans, as well as a thick condiment called 
 "tao tjiung," similar to the Japanese "miso." According to Prinsen- 
 Geerligs, these preparations have the following composition : 
 
 'Chem. Ztg., 10 (1886), p. 140. 
 
 2 Auier. Chem. Jour., 16 (1894), p. 49. 
 
 3 Compt. Rend. Acad. Sci. Paris, 90 (1890), pp. 1177-1180. 
 ^Loc. cit. 
 
 6 Chem. Ztg., 20 (1896), pp. 67-69. 
 
36 
 
 Table 11. — Composition of Chinese soy-bean preparations. 
 
 Tao hu (bean cheese) 
 
 Milk from boiled soy beans. 
 
 Tao yu (soy sauce) 
 
 Tao tjiung" 
 
 "Water. Protein. 
 
 Percent. 
 76.15 
 93.10 
 57.12 
 62.86 
 
 Percent. 
 
 la. 15 
 3.13 
 7.49 
 
 12.67 
 
 ^ ^ Carbohy- 
 ^**- ' drates. 
 
 Per cent. 
 7.09 
 1.89 
 
 Per cent. 
 1.40 
 
 16.03 
 
 6 13. 78 
 
 Ash. 
 
 I TTndeter- 
 ! mined. 
 
 Per cent. 
 
 2.21 
 
 0.51 
 
 a 18. 76 
 
 6.71 
 
 a Including 17.11 .sodium chlorid. 
 
 blncludini; 3.78 crude fiber. 
 
 The Japanese preparation made from soy beans, similar to those 
 mentioned above, have been described by a number of investigators. A 
 brief account of theiu has been given in a previous publication of this 
 Department.' As shown by their composition, these soy-bean prepara- 
 tions contain a high percentage of nutrients. Tliey are eaten in large 
 quantities and form important articles of diet. A number of diges- 
 tion experiments have been reported in which the Japanese ])repara- 
 tions formed a considerable part of the diet, and on the basis of the 
 results obtained the preparations are considered to be very well 
 assimilated. 
 
 PHASEOLUS. 
 
 The seeds of two varieties of Phascoins mnngo are largely used by 
 the Chinese in San Francisco. One of these, known as "iuk tau," 
 designated by the characters |^_a^, is a small green bean. The individ- 
 ual seeds weigh only about 0.044 gram. They are slightly flattened at 
 the ends and have a rather long hilum. Plants were grown at Berkeley, 
 Cal., from these seeds and yielded a good return, though the plant 
 was not as prolific as the soy bean. The appearance of one of them is 
 shown in PI. VIII. The analyses in the table below show that these 
 beans differ in composition but slightly from the commonly cultivated 
 varieties of P. vnh/aris. 
 
 The Chinese use these beans largely in the preparation of "bean 
 sprouts," which are simply seeds that have been soaked in water and 
 allowed to germinate till the young plants are severaliuches in length. 
 This product is said to be used in making soup. It is difficult to 
 imagine what advantage the young plants possess over the original 
 seeds for such a ])urpose, unless they impart a different flavor. From 
 an economic standpoint the process is wasteful, as it involves the trans- 
 formation of albuminoids into various cleavage products and amido- 
 compounds, whose nutritive value is thought to be but slight. Henry^ 
 refers to the preparation of a kind of vermicelli from this bean, but 
 such a i)reparation was unknown in San Francisco. 
 
 The second form of Phaseolus in use Is of a dull red color, rather 
 larger than the previous variety, and more nearly spherical in shape. 
 This is known as the "huang tau" or red bean, designated by the 
 characters -fj^. Kelluer analyzed the seeds of a variety of P. radiatus 
 
 ' U. S. Dept. Agr., Farmers' Bui. 58, Appendix. 
 
 * Notes on the Economic Botany of China, p. 13, Shanghai, 1893. 
 
U. S. Dept. of Agr,, Bui. 68. Office of Expt. Stations. 
 
 Plate VIII. 
 
 Plant of Phaseolus mungo. 
 
37 
 
 grown in Japan, wliicli is very probably the same as the one here 
 described, as the two species are by some authors regarded as identical. 
 Harrison and Jenmau report the analysis of mungo beans grown in 
 British Guiana. 
 
 The composition of the red and green P. mungo is shown in Table 12. 
 For purposes of comparison the composition of a green form of P. mungo 
 reported by Church, P. radiaitis' reported by Kellner, and "mungo" 
 beans reported by Harrison and Jenman, are also given. 
 
 Tablk 12. — Composition of seed of Phaseolus iipp. 
 
 Phnseolus mungo {green) : 
 
 Original materia/ 
 
 Water- free substance 
 
 Fhaseolus mungo (red) : 
 
 Original iii;iterial 
 
 Wat er-lree substance 
 
 Ph aneolus mungo 
 (ureen) : a 
 
 Oriiiiiial material 
 
 "Water-freo substance 
 
 Phaseolus radiatut 
 (red) :c 
 
 Original material 
 
 Water-free substance 
 
 "Mungo beans:" e 
 
 Original material 
 
 Water- free substance 
 
 Per ct. 
 8.83 
 
 Per ct. 
 22.64 
 24.83 
 
 21.06 
 23.52 
 
 22.2 
 24.9 
 
 18.30 
 20.84 
 
 » 
 
 ■:^ 
 
 
 
 
 
 c 
 
 
 
 ■= 2 
 
 
 
 c 
 
 X S 
 
 
 
 ^ 
 
 c*^ 
 
 
 
 ■< 
 
 < 
 
 Per ct. 
 
 Perct. 
 
 21.88 
 
 0.76 
 
 24.00 
 
 .83 
 
 18.19 
 20.32 
 
 2.87 
 3.20 
 
 16.74 
 19.06 
 
 20.54 
 24.09 
 
 1.56 
 1.78 
 
 1.52 
 1.78 
 
 Per ct. 
 1.34 
 
 1.47 
 
 2.7 
 3.0 
 
 1.42 
 1.62 
 
 1.96 
 2.29 
 
 Per ct. 
 48.54 
 53.23 
 
 48.36 
 54.02 
 
 57.40 
 65.38 
 
 /6.71 
 
 7.87 
 
 P.ct. 
 0.00 
 
 .00 
 
 .00 
 .00 
 
 6 54.1 
 b60.7 
 
 d2. 02 
 (?2. 31 
 
 P.ct. 
 
 4.52 
 4.95 
 
 5.02 
 5.61 
 
 5.8 
 6.5 
 
 3.74 
 4.38 
 
 P. 
 
 2.60 
 2.96 
 
 3.39 
 3.98 
 
 P.ct. 
 11.28 
 11.39 
 
 9.61 
 10.73 
 
 aEeported by Churcli, Pood Grains of India. London, Chapman and Hall, 1886, p. 151. 
 6 Carbohydrates by difl'erence. 
 
 cReported by Kellner, Landw. Vers. Stat., 30 (1881), p. 42. 
 d Determined by ditiereuce. 
 
 e Reported by Harrison and Jenman, Rpt. A jr. Work Bot. Gard. British Guiana, 1891-92, p. 73. 
 /The authors also report in the fresh substance 41.23 per cent digestible liber and 1.80 per cent pec- 
 tose, gums, etc. 
 
 Harrison and Jenman^ also report analyses of a number of beans 
 whose botanical names were not known. They were called red " mote," 
 "Chinese bean," '' kingto," and " octow." 
 
 The results of Kellner's analysis, as well as those secured at San 
 Francisco (the red Phaseolus mungo), show that the two varieties 
 have essentially the same composition. The analyses also show that 
 about 90 jjer cent of the protein is in the form of albuminoids. Osborne^ 
 has studied the proteids of the white-podded adzuki bean, which is 
 presumably the same variety as the one analyzed by Kellner. He 
 found they consisted of phaseolin and a hitherto unknown globulin. 
 
 DOLICHOS. 
 
 Still another bean, which is largely used by the Chinese of San Fran- 
 cisco, is Dolirlios sesquipedaUs. These seeds are white, with a promi- 
 nent black ring which surrounds the hilum. The beans weigh about 
 0.31 gram each. The skin covering the seed is especially tough, but 
 
 ' Loc. cit. 
 
 2 Jour. Amer. Cheui. Soc, 19 (1897), p. 509. 
 
38 
 
 can readily be removed after soaking in water. The beans are most 
 largely used in the form of " bean sprouts " after the removal of this 
 skin. They are known as ''mi tau" (/g ji); but this name and its 
 accompanying characters have not been found in any of the books on 
 Chinese botany now accessible to the author. The plant is said to be 
 a native of South America, and was probably recently introduced into 
 China. As shown below, the composition of the dry bean differs but 
 slightly from that of our commonly cultivated variety of beans. 
 
 The green pods of this species are also largely used as a snap bean. 
 These are from 50 to 70 centimeters in length, and contain from 10 to 16 
 seeds. They are more slender than most of our string beans, and 
 slightly ridged along the middle of the two valves. This vegetable is 
 known as " tou kok," and is cultivated by the Chinese along the Sac- 
 ramento Eiver, arriving in market early in July. It is said that it has 
 come into use among the white residents of the central portion of Cali- 
 fornia, and is considered a valuable variety of string bean. The plant 
 is somewhat tender, and requires a long season for development, but 
 Bailey ^ reports fair success with it in the eastern United States. 
 
 Table 13 shows the composition of the seeds and green pods of D. 
 sesquipedalis as compared with that of D. lablab, cowpeas, and other 
 beans. 
 
 Table 13. — Composition of Doliclios and other heans. 
 
 
 
 S 
 o 
 
 2 
 '3 
 
 a 
 
 a 
 .=1 
 
 .a o 
 
 1^ 
 
 i 
 
 i 
 
 fcJC 
 
 s 
 
 M . 
 
 'S ej 
 
 u 
 
 g 
 
 a 
 
 < 
 
 a 
 a 
 
 
 
 Dolichog sesqnipedalis 
 (seed) : Per ct. 
 Original mat eriaL... 10.98 
 Water-free substance: 
 
 Per ct. 
 22. U 
 25.54 
 
 24.4 
 
 27.8 
 
 21.4 
 24.6 
 
 22.5 
 25.74 
 
 4.54 
 22.63 
 
 2.3 
 21. 30 
 
 Per ct. 
 17.84 
 20. 04 
 
 P.ct. 
 4.90 
 5.50 
 
 P.ct. 
 2.66 
 2.99 
 
 1.5 
 1.7 
 
 1.4 
 1.61 
 
 1.8 
 2.07 
 
 .53 
 2.04 
 
 .3 
 2.77 
 
 Per ct. 
 
 45.91 
 51.57 
 
 Per ct. 
 
 4.27 
 4.80 
 
 Per ct. 
 
 0.00 
 
 .00 
 
 Per ct. 
 2.80 
 3.15 
 
 1.2 
 1.4 
 
 4.1 
 4.71 
 
 4.4 
 5.03 
 
 2.56 
 12.74 
 
 1.9 
 17.59 
 
 P.ct. 
 2.86 
 3.21 
 
 3.4 
 ?.9 
 
 3.4 
 3.91 
 
 3.5 
 
 4.0 
 
 1.17 
 5.81 
 
 .8 
 7.41 
 
 P. Ct. 
 
 7.78 
 8.74 
 
 Doliclios lahlah (seed) -.a 
 
 Original material 12. 1 
 
 &57.8 
 6 65. 8 
 
 6 56.7 
 b 65. 17 
 
 h 55. 2 
 b 63. 16 
 
 
 
 
 
 Cowpeas ( Vigna catjang) 
 (Average of 13 analy- 
 ses) :C 
 
 13.0 
 12.6 
 
 79.92 
 
 SB 9 
 
 
 
 
 
 
 
 
 Dried beans (average of 
 11 analyses) :d 
 
 
 
 
 
 
 
 
 
 3.04 
 15.14 
 
 1.50 
 7.49 
 
 
 Dolichos sesqvipedalit 
 (green pods) : 
 
 ( )riginal material 
 
 Water- free substance 
 
 2.74 
 13. 66 
 
 1.75 
 
 8.72 
 
 3.26 
 16.24 
 
 3.53 
 17. 56 
 
 String beans (average of 
 5 analyses) : d 
 
 6 5.5 
 b 50. 93 
 
 
 
 
 
 
 
 
 
 
 
 a Reported by Church, Food Grains of India. London: Chapman and Hall, 1886, p. 161. 
 
 b (,'arbohydrates by ditierence. 
 
 c r. S. Uept. Agr., (XHce of Experiment Stations Bui. 28 (rev. ed.), p. 37. 
 
 ri Ibid., p. 65. 
 
 Harrison and Jenman ^ report the composition of two varieties of 
 D. lahlah, the crinkled-podded Bouavist bean and the flat-podded 
 
 ' New York Cornell Sta. Bui. 67, p. 197. 
 
 -Rpt. Agr. Work Bot. Gard. British Guiana, 1891-92, p. 70. 
 
39 
 
 Boiuivist bean. Their results are not quoted, since the beans were 
 evidently a different species from those purchased in San Francisco. 
 
 Both fresh and dry, the Bolichos analyzed in this laboratory compares 
 favorably with the ordinary American beans, and were found on trial 
 to be a satisfactory substitute for these, provided only the tough skin 
 was removed from the dry seeds, which was readily effected. This 
 species seems worthy of a general trial. 
 
 VARIOUS WATER PLANTS. 
 
 Nelumhium speciosum. — In reviewing the numerous uses of the sacred 
 lotus (p. 17), it was stated that the seeds were an article of food among 
 Asiatic nations, and it is not surprising to tind them on sale in San 
 Francisco. Their peculiar shape, the remarkably mature condition of 
 their embryos, and the peculiar form of the embryo itself, all render the 
 seed easy of recognition. Two forms were obtained, the one being 
 somewhat larger and more irregular in shajie than the other, but it was 
 impossible to ascertain whether they were the product of plants which 
 differed materially from each other. The smaller and more commonly 
 used of the two forms is known as "pak lin" and is designated by the 
 characters j^j^; the larger is known as " seung lin" ('^j||). In all 
 instances the thick outer coating of the seed on sale in San Francisco 
 had been removed, leaving only the inner kernel. Under favorable 
 conditions of temjierature and moisture the seeds purchased germinated 
 rapidly, and in the course of a few months from the time of starting pro- 
 duced thrifty plants. Thus far the i:>lants from the two varieties of 
 seeds do not diner materially from each other nor from the commonly 
 cultivated form of lotus. As yet none have produced flowers. The 
 germination of the seeds is somewhat interesting from a botanical 
 point of view and is figured in PI. II, fig. 4. 
 
 The seeds are eaten by the Chinese either raw, boiled, or roasted, 
 being used as we use chestnuts. They are also said to be used in soup, 
 though this seems to be a general term used to designate any mixture 
 that has been boiled. The author w-as also assured that a form of 
 "arrowroot," i. e., starch, is made from the seed. 
 
 The dark green germ is decidedly bitter and must be removed before 
 the seed becomes j^alatable. Its bitterness has given rise to the Chinese 
 saying, "bitter as the plumule of the lotus seed." 
 
 Analyses of the two forms are given in Table 14, p. 41. In the sam- 
 ple of large seed the germs, which form 1.3 per cent of the weight 
 of the entire seed, were removed before analysis, but in the case of the 
 smaller form the entire seed was analyzed. 
 
 An inspection of the two analyses shows that the seeds contain a 
 high percentage of nutrients. The protein content is considerably 
 above that of the commonly cultivated cereals, but does not approach 
 that of the leguminous seeds. Nearly all of it is of an albuminoid 
 nature. As might have been expected, starch is the most important 
 nonuitrogeuous constituent, though small amounts of other carbohy- 
 
40 
 
 drates are also present. The starch found in both varietiea* differs in 
 form from that found in the roots of the lotus. 
 
 Uuryaleferox. — Tins is another aquatic, closely related to the lotus 
 both in habit of ^ixowth and in botanical aftiuities. It is designated 
 by the characters -^i^. Lite the lotus its seeds form a part of the com- 
 plex dietary of the Asiatic races.^ These seeds were occasionally met 
 with in the Chinese stores, but only in a broken and imperfect condition, 
 apparently the result of a mechanical process intended to crack them 
 open. Xevertheless the broken seeds showed such a remarkable resem- 
 blance to those of our native l^iqiliar pohjsepalum^ which also furnishes 
 seeds used by the Klamath Indians of Oregon- for food, that the 
 author was at once led to suspect their relationship, and a dictionary 
 in the Cantonese dialect, by John Eitel, associates the character used 
 for them with the name given. The seeds are much smaller than those 
 of the lotus, more larinaceous, and their embryos show only a moderate 
 degree of development. The characters used to designate them are quite 
 different from those used for the lotus seeds and the name commonly 
 used is '-tsz shat," which, however, does not appear in any of the 
 works descriptive of Chinese botany accessible to the author. They 
 are used principally as an addition to soup, in much the same manner 
 as we use barley, but are also said to be used for the manufacture of 
 starch. 
 
 The accompanying analysis (Table 14, p. 41) shows tliat they possess 
 a food value about equivalent to that of the cereals, the principal con- 
 stituent being starch, whicli is present in the form of exceedingly small 
 grains, as shown in PI. Ill, fig. 5. Sugars are probably not present, 
 as the small amount reported in the analysis may represent traces of 
 starch which repeated filtration failed to remove from the aqueous 
 extract of the seeds. But little information is accessible regarding 
 adaptability of the plant to the conditions found in the United States. 
 It is already in cultivation in the North Atlantic States and other 
 regions as an ornamental aquatic. 
 
 Trapa hispinosa. — The curiously-shaped seeds of this plant, which 
 may be aptly compared with the head and horns of a cow, are familiar 
 objects in the Chinese bazaars, where they are sold as curiosities. The 
 regular Chinese merchant, however, is quite sure to have them in stock 
 during the spring mouths, and they are sold by him as an article of 
 diet. In the bazaars they are known as the "horn chestnut." The 
 Chinese name is "ling ko" (^H). This is also the name by which they 
 are known in the Chinese classics, as the seeds are sometimes included 
 among the five food grains of China. The seeds do not keep well, and 
 those which find their way into the bazaars are often entirely decayed 
 within. 
 
 The fresh seeds have a kernel which, in consistency and taste, resemble 
 the chestnut. They will germinate readily if placed in a vessel of water 
 
 ' Bretscluieider, Jour. China Branch Roy. Asiatic. Soc, 25 (18 90), p. 218. 
 •^Coville, Contrib. U. S. Nat. Herb., 5 (1897), No. 2, p. 96. 
 
41 
 
 and i^roduce the curious floating foliage characteristic of this genus of 
 plants. (See PI. II, tig. 1, and PI. Ill, tig. 4.) 
 
 Authors are at variance as to the correct name of the plant produc- 
 ing these seeds. Forbes and Hemsley ' regard it as one of the forms 
 of the more widely known European T. natans inasmuch as they find 
 many intermediate forms between the Cliinese and the European species. 
 By other writers the plant is variously designated as T, hlspinosa^ 
 T. bicornis, T. cochinchinensis, and T. incisa. The plants to which these 
 terms are applied are probably identical. As a means of distinguish- 
 ing it from T. natans, from which the seeds analyzed by the author 
 certainly ditfer, the name T. hisphiosa has been adopted in this report. 
 
 The composition of the two sorts of Nelurabium seed, of Euryaleferox, 
 and Trajm hispinosa are shown in the following table: 
 
 Table 14. — Composition of seeds of various water plants. 
 
 
 1 
 
 Protein. 
 Albuminoids. 
 
 -Si 
 
 ^ 
 
 .a 
 
 
 .2 2 
 
 IS 
 
 1 
 
 
 •6 
 
 .a 
 
 5 
 
 Nelumhium speciosum 
 (large form) : 
 
 Original material . . . 
 
 Water-free substance 
 Nelutnbium gj^eciosum 
 (smaller form) : 
 
 Original material ... 
 
 Water-free s ubstance 
 Euryaleferox: 
 
 Original material . . . 
 
 'Water-free .substance 
 Trapa bispinofa: 
 
 Original material . . . 
 
 Water- free substance 
 
 Per ct. 
 8.72 
 
 9.40 
 13.10 
 10.59 
 
 Peret. P»ret. 
 16.64 : 15.47 
 18.23 16.95 
 
 17. 73 17. 64 
 19.57 19.46 
 
 9. 79 • 9. 79 
 11.26 11.26 
 
 10.88 10.42 
 12. 16 1 11. 65 
 
 Perct. 
 1.17 
 1.28 
 
 .09 
 .10 
 
 .46 
 .51 
 
 Per ct. 
 2.44 
 2.67 
 
 2.96 
 3.26 
 
 4.90 
 5.64 
 
 .65 
 .73 
 
 Perct. 
 
 51.64 
 56.57 
 
 40.63 
 
 44.84 
 
 68.07 
 78.33 
 
 60.39 
 67.53 
 
 Perct. 
 4.09 
 4.48 
 
 9. .55 
 10.54 
 
 1.59 
 1.83 
 
 P.ct. 
 2.41 
 2.64 
 
 3.95 
 4.42 
 
 P. ct. 
 3.15 
 3.45 
 
 2.95 
 3.26 
 
 .83 
 .96 
 
 1.41 
 1.58 
 
 P.ct. 
 3.03 
 3.32 
 
 4.15 
 4.58 
 
 .54 
 ,62 
 
 2.57 
 2.87 
 
 p.ct. 
 
 7.88 
 8.63 
 
 12.63 
 13.94 
 
 1.18 
 1.36 
 
 9.56 
 
 10.70 
 
 The analysis here reported of the seeds of Trapa hispinosa is not 
 typical as regards water content, since the seeds used had been kept for 
 a considerable time before they were analyzed and were unusually dry. 
 The average weight of the seeds is about 4 grams and in those used the 
 kernel constituted about 64.3 per cent of the total. Their food value is 
 slightly less than that of the seeds of the Euryale, but, judged solely 
 by their composition, they nevertheless contain a high percentage of 
 nutritive material. 
 
 CHINESE MILLET. 
 
 Though this grain was not found on sale in San Francisco, it has a 
 widely extended use in China and Japan. The sample analyzed in this 
 laboratory was given to the author by Professor Fryer, who states that 
 it is more especially used in the northern part of China. He considers 
 it a very valuable article of diet and believes this plant should be intro- 
 duced into cultivation in the United States. The species (presumably 
 Paniciim) to which the sample belongs can not be determined until 
 
 ' Index Flora} Sinensis, 1887, p. 311. 
 
42 
 
 plants have been grown from it. Several species of Pgftiicnm and 
 closelj" related genera are in cultivation in China. Church reports the 
 comj)osition of a considerable number of millets used as food in India, 
 and discusses their food value at some length. 
 The comi)ositiou of the sample is shown in the following table: 
 
 Table 15. — Composition of millet. 
 
 
 
 
 to 
 
 5ti 
 
 
 
 
 M 
 
 
 
 -i 
 
 
 
 
 2 
 '3 
 a 
 
 r3 . 
 
 
 
 be 
 
 to . 
 
 
 
 B 
 
 
 
 a 
 'S 
 o 
 
 a 
 
 • 1-1 <s 
 
 
 
 m 
 
 P 
 
 •5 
 
 3 
 
 ^ 
 
 S 
 S 
 
 
 ^ 
 
 ^ 
 
 <) 
 
 fa 
 
 m 
 
 O 
 
 W 
 
 
 < 
 
 t> 
 
 Panicum sp. : 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 Per ct. 
 
 P. ct. 
 
 P.ct. 
 
 P.ct. 
 
 P.ct. 
 
 Original material 
 
 11.60 
 
 8.56 
 
 8.08 
 
 0.48 
 
 2.64 
 
 67.47 
 
 0.06 
 
 0.00 
 
 3.32 
 
 1.79 
 
 4.56 
 
 VV ater-l'ree substaii ce: 
 
 9.68 
 
 9.14 
 
 .54 
 
 2.94 
 
 76.32 
 
 .07 
 
 .00 
 
 3.75 
 
 2.02 
 
 5.16 
 
 Panicum italicum: a 
 
 
 
 
 
 
 
 
 
 
 
 Orisiual material 12. 04 
 
 7.40 
 
 6.64 
 
 .76 
 
 3.87 
 
 45.73 
 
 628.48 
 
 1.37 
 
 1.11 
 
 
 Water-free substance 
 
 Panicum miliaceum: c 
 
 8.43 
 
 7.56 
 
 .87 
 
 4.40 
 
 51.99 
 
 632 
 
 38 
 
 1.54 
 
 1.26 
 
 
 
 
 
 
 
 
 
 
 
 
 Original material 12. 
 
 12.6 
 
 
 
 3.6 
 
 
 (i69.4 
 
 
 1.0 
 
 1.4 
 
 
 14.3 
 
 
 
 4.1 
 
 
 d78.9 
 
 
 1.1 
 
 ].6 
 
 
 
 
 
 
 
 aTteported by Kellner, Landw. Vers. Stat., 30 (1884), p. 42. 
 
 6 Determined by ditlereuce. 
 
 c Reported by Church, loc. cit., p. 43. 
 
 d Carbohydrates by difference. 
 
 For the sake of comparison, two analyses, one of a Japanese- grown 
 sample, the other of a sample grown in India,' are included in the table 
 above. All three analyses confirm the statement that in composition 
 this grain ranks with the more commonly cultivated cereals. It is pos- 
 sible that it might find an important place among our cultivated grains. 
 
 FRUITS, NUTS, AND FLOW^ERS. 
 
 Nepheliuni litchi. — This tree furnishes the well known "li chee" or 
 "lai chi" nuts (^;f^), which are so frequently used as tokens of good 
 will by the Chinese laundryman or vegetable peddler about the time 
 of their New Year. The edible portion of the li chee nut is really the 
 fleshy aril which immediately surrounds the smooth brown seed and is 
 in turn surrounded by the thin warty shell of the fruit itself. The 
 nuts may be obtained in the dry form and also preserved with sugar 
 in cans. In the sample analyzed the aril formed 40 per cent of the 
 weight of the dry seed, but these figures can scarcely be considered 
 typical as the fruits were kept for some months before being used. 
 The composition of "li chee" nuts is shown in Table 10, p. 45. As 
 might be expected from their sweet taste, sugar, especially reducing 
 sugar, is their most important constituent. Other nutritive substances 
 are present in small amounts. 
 
 The tree has been a favorite subject for greenhouse treatment in 
 temperate climates for many years and is also reported to tolerate the 
 
 * Food Grains of India, pp. 34-59. London, Chapman «fc Hall, 1886. 
 
43 
 
 climatic conditions found in southern California in a satisfactory- 
 manner. 
 
 Nephelium longan. — The fruit of this tree, designated by the char- 
 acters HB^gJ, differs from that of N. litchi in beiug somewhat smaller 
 and smooth, otherwise it closely resembles the "li chee" nut. How- 
 ever, it is considered inferior to the li chee by the Chinese. The sample 
 was kept for some time before analysis, and hence the figures for water 
 content and those based on this determiuation are scarcely typical. 
 The aril formed 40 per cent of the entire seed. Its composition is 
 shown in Table 16, p. 45. In composition it does not differ greatly 
 from the "li chee" save in the relative amounts of reducing and cane 
 sugars present. 
 
 Zizyphus jujuha. — This is the Indian fig largely cultivated in trop- 
 ical Asia, Africa, and Australia, and is designated by the characters 
 ifiH^- ^^^ dried fruits are used by the Chinese in the same manner 
 as the two mentioned above and are also well liked by Europeans. 
 The European species [Z. vulgaris), is, however, better known and fur- 
 nishes the important constituent of the jujube paste of the confec- 
 tioners' shops. The composition of the fruit of Z. jujuha is shown 
 in Table 10, p. 45. 
 
 Ganarium album. — The so-called Chinese olive is the fruit of several 
 species of Canarium, C\ album and G.pimela being probably the two 
 species whose fruit is chiefly used. The Chinese olive is a fleshy drupe 
 3 to 6 centimeters in length, which contains a hard, triangular, sharp- 
 pointed seed. (See PI. II, fig. 6.) Within this are found one or more oily 
 kernels. The fruits, known as "pak lam" (|^^), are found on sale in 
 San Francisco, either green or salted and dried. They are also sold 
 in the form of certain highly flavored preparations which are not 
 generally acceptable to the American palate. The flesh of the fresh, 
 yellowish-green fruit, like that of the true olive, is somewhat acrid and 
 disagreeable, and requires special treatment before it can be made pal- 
 atable. Smith ^ says that the fruits are often preserved in salt, and 
 are also added to wine to moderate or counteract its effect. 
 
 The sample analyzed by the author (see Table IG, p. 45) represents the 
 fresh pulp of the fruit. The most important constituent is fat, which 
 forms nearly one-fourth of the total nutritive material. 
 
 The kernel of the seed of one variety of Chinese olive is also sold 
 under the name of "lam yen" (/f^fn). The peculiar form of these seeds 
 is shown in PI. II, fig. 7. They are covered with a thick, reddish skin, 
 but have the consistency and general characteristics of many other oily 
 seeds. Microscoi)ical examination showed the presence of well-formed 
 aleurone grains, the crystalloids of which are remarkably well developed. 
 The analysis of the seeds (reported in Table ]6) shows that in comi^osi- 
 
 ' Contributions Toward the Materia Medica and Natural History of China, p. 50. 
 Shanghai, 1871. 
 
44 
 
 tion they closely resemble ■walnuts and similar nuts. Thefiat present 
 consists of a yellow fluid oil, which absorbs 83.17 per cent of iodin. 
 AVhen fresh, the kernels are as palatable as walnuts and other common 
 nuts. Little is known as to the adaptability of Chinese olive trees to 
 other than tropical climates. It is questionable whether many j)ortions 
 of the United States would furnish the conditions favorable to their 
 growth. 
 
 Ginlcgo hiloha {SaUsburia adiantifolia). — A detailed account of this 
 tree seems scarcely necessary, since it is well known — by nurserymen, 
 at least — throughout the United States, though its merits as an orna- 
 mental tree are by no means fully appreciated. The trees grow readily 
 and have fruited abundantly as far north as Washington, D. C. The 
 white, thin-shelled nuts (PI. II, fig. 2), known as "pak ko" or "gink 
 ko" (g|g), are eaten by the Chinese, being first roasted or boiled, like 
 chestnuts. They are somewhat acrid in the fresh state, and have a 
 very disagreeable odor. When roasted or boiled, their flavor is peculiar. 
 Tbey are said by Hanbury,^ Smith, and others to assist in digestion and 
 to be eaten after meals for this purpose by the Chinese. 
 
 The pulp surrounding the seed of the fresh fruit has been examined 
 by Bechanip,- who obtained from it the entire series of homologous fatty 
 acids, from formic acid to caprylic acid. He also noted the occurrence 
 of starch in the endosperm, though a complete analysis of the seed does 
 not seem to have been made. The nuts found in San Francisco were 
 evidently unusually dry. The analysis (Table IG, p. 45) shows that 
 by far their most important constituent is starch. This is somewhat 
 remarkable, as it is the only instance of which record has been found 
 in which starch has been observed in any considerable quantity in the 
 seed of a coniferous plant. 
 
 Some experiments were conducted by a student in this laboratory to 
 determine whether the seed contained a digestive ferment. Samples 
 of fish flesli were treated under suitable conditions with a decoction 
 of the seeds, but no digestive action was observed. 
 
 Hemerocallis fulva. — Still another and very unusual vegetable sub- 
 stance largely used as a flavoring ingredient by the Chinese consists of 
 the dried flowers of II. fulva., the day lily of American gardens. This 
 substance is known as "-kam cham t'soi/' or the "gold-needle vege- 
 table," and is designated by the characters '^%\'^. It is reported by 
 Bretschueider ' to be in common use. As noted on page 20, Davy* 
 refers to the use of the flowers of Xi7i»m hulhiferum and H. (jraminca as 
 food by the Chinese. The composition of flowers of H. fulva is shown 
 in Table 16, p. 45. 
 
 ' Notes on Chinese Materia Medica, Scientific Papers, 1876. 
 
 ^ Compt. Keud. Aca<l 8ci. Paris, 58 (1864), p. 135; Ann. Cliini. etPliys.,4. ser., 1 (1864), 
 p. 288. 
 
 ^'.lour. China Branch Roy. Asiatic Soc, 15 (1880), p. 110. 
 ••Erythea, 6 (1898), p. 25. 
 
45 
 
 When judged by their composition, flowers of H. fulva are seen to pos- 
 sess a rather high food value. They are used, however, rather as a 
 condiment than as an article of diet. 
 
 The composition of the fruit of N'ephelium litchi ("li chee" nuts), N. 
 longan, Zizyphus jujuba, Canariiim album, (Chinese olive), pulp and 
 seeds; Ginl-go biloba, "gink ko" nuts, and the dried flowers of Henie- 
 rocalUs fulva are shown in the following table: 
 
 Table 16. — Composition of fruits, nuts, and flowers. 
 
 
 i 
 
 'S 
 1 
 
 'o 
 
 a 
 
 B 
 % 
 
 Cm 
 
 
 .a 
 u 
 
 1 
 
 1/2 
 
 08 
 W) 
 3 
 m 
 
 (D 
 
 a 
 a 
 O 
 
 s 
 .9 £ 
 
 5 
 
 < 
 
 '6 
 « 
 
 .3 
 
 a 
 
 .2 
 
 (D 
 ri 
 
 a 
 P 
 
 Nephelium litchi (aril) : 
 
 Perct. 
 
 Per ct. 
 2.91 
 3.43 
 
 5.01 
 5.63 
 
 2.93 
 3.39 
 
 .77 
 2.86 
 
 16.44 
 17.44 
 
 5.90 
 11.21 
 
 10.11 
 11.99 
 
 Peret. 
 
 P.ct. 
 
 Per ct. 
 1.44 
 1.69 
 
 1.04 
 
 1.17 
 
 Perct. 
 
 Per ct. 
 
 4.47 
 5.25 
 
 37.50 
 42.11 
 
 13.06 
 15.09 
 
 .53 
 1.99 
 
 .92 
 .97 
 
 3.58 
 6.79 
 
 30.51 
 36.19 
 
 Per ct. 
 66.58 
 
 78.27 
 
 27.34 
 30.70 
 
 42.19 
 48.74 
 
 1.95 
 7.26 
 
 .00 
 .00 
 
 .00 
 .00 
 
 12.40 
 14.71 
 
 p.ct. 
 
 4.15 
 15.48 
 
 3.20 
 3.39 
 
 .88 
 1.68 
 
 8.74 
 10.37 
 
 P.ct. 
 2.21 
 2.60 
 
 2.31 
 2.59 
 
 1.73 
 2.00 
 
 1.50 
 5.61 
 
 5.16 
 
 5.47 
 
 2.00 
 3.81 
 
 3.64 
 4.32 
 
 P.ct. 
 
 7 45 
 
 
 j 
 
 8 75 
 
 Nephelinm longan (aril) : j 
 
 Original material 10. 94 
 
 Water-free substance 
 
 Zizyphus jiijuba (])ulp) : 
 
 
 
 15 86 
 
 
 
 17.80 
 26 65 
 
 
 
 
 
 
 30 78 
 
 Canarium album (pulp) : 
 
 Original material 
 
 Water-free substance 
 
 Canarium sp. (seeds) : 
 
 Original material 
 
 Waterfreesubstance 
 
 Ginlgo biloba (seeds) : 
 
 Original material 
 
 Water- free substance 
 
 Hemerocallis fulva (AviaA 
 flowers) : 
 
 73.22 
 
 5.71 
 
 47.34 
 
 15.70 
 
 .61 
 2.28 
 
 10.44 
 17.44 
 
 5.18 
 9.84 
 
 .15 
 .57 
 
 .00 
 .00 
 
 .72 
 1.37 
 
 6.55 
 24.46 
 
 ,59. 57 
 63.18 
 
 .81 
 1.53 
 
 3.42 
 
 3.16 
 11.79 
 
 .00 
 .00 
 
 33.90 
 64.34 
 
 8.17 
 30.53 
 
 9.00 
 9.54 
 
 5.58 
 10.60 
 
 9.50 
 
 
 
 
 4. 06 7. 09 
 
 11.27 
 
 
 
 
 
 
 
 FUNGI AND ALG-ai. 
 
 Russiila sp. — Several species of fungi are largely used in the Chinese 
 dietary, and when dried are standard articles of trade in San Francisco. 
 The one most largely used is a white-spored species of Agaricini, prob- 
 ably of the genus Eussula. It is designated by the characters ^jj^. 
 It is easily distinguished from the American edible species by the 
 tough, coriaceous character of both pileus and stipe, the latter being 
 short and eccentrically attached. In color it is yellow, except the top 
 of the pileus, which is brown or purple. At least two other species of 
 this group are on sale in San Francisco, but these are delicacies rather 
 than staple articles of diet, some of them retailing as high as $2 per 
 pound. 
 
 The composition of the liussula is shown in Table 1 7, p. 47, and pre- 
 sents some interesting features. The figures for protein are far below 
 those which have been assigned to edible fnngi of this type, being only 
 one- third to one-half the amounts reported for Agaricns campestris, 
 
46 
 
 species of Tuber,^ etc. Of the protein, about four-fifths is m the form 
 of albumiuoids. 
 
 The exact nature of the large amount of carbohydrates present is 
 unknown, and would form an interesting subject for investigation. 
 
 Peziza auricula. — Most of the notes of the earlier writers on Chinese 
 botany contain references to the use of a peculiar mushroom which 
 grows on decaying wood. This is a species of Peziza, probably 
 P. auricula. It is designated as " mo eh" (Loureiro) or " mu ery " (yf;;^), 
 by Henry, the English equivalent of which would be " wood ears," and 
 is used in large quantities in San Francisco. The composition of the 
 fungus is shown in Table 17, p. 47. The protein content of this fungus is 
 remarkably low; sugars seem to be entirely absent, but some starch- 
 like substance of an unknown nature forms more than one-third of the 
 total weight. The fungus is decidedly tough and woody, and appar- 
 ently possesses no especially valuable food characteristics. 
 
 Nostoc commune flagelliforme. — Of the algae, one form which occurs 
 in black tangled masses resembling horsehair is largely used by the 
 Chinese. This was identified by Prof. W. A. Setchell as Nostoc com- 
 mune flageUlforme. It is designated by the characters ^^. The use 
 of this plant as an article of food does not seem to have been hereto- 
 fore reported. Treated with boiling water, it forms a gelatinous mass, 
 and is used as a thickening medium for various culinary preparations, 
 one of the most esteemed being a combination of this vegetable with 
 dried shrimps. The analysis shows a high percentage of protein, but 
 only insignificant amounts of other food constituents. 
 
 Other forms of alga- which were identified by Professor Setchell are 
 Forplii/ra perforatCj and P. nereocystis, both of which are collected in 
 large amounts at the Bay of Monterey, and two other species of the 
 same genus, presumably P. suborbiculata and P. tenera, which are cul- 
 tivated in Japan. 
 
 Sj)ecies of Porphyra are used in considerable quantities for food by 
 the natives in southeastern Alaska. The alg;e are collected, pressed 
 together and dried, and when used, are shaved into warm water and 
 boiled to a thick jjorridge.^ A sample of this alga' analyzed in the 
 chemical laboratory of the Connecticut (Storrs) Station, had the compo- 
 sition reported in Table 17, p. 47. 
 
 The composition of the fungi and alga^ analyzed, together with 
 analyses of fungi quoted for purj)Oses of comparison, is reported in the 
 following table : 
 
 1 Kitnig, Cliemie der mensclilichen Nahrungs- imd Genussmittel, 3. ed., I, p. 747. 
 -This informatiou is furnished by W. H. Evans, Ph. D., of this Department. 
 
47 
 
 Table 17. — Compontion of fungi and algw. 
 
 
 
 
 
 !2 
 '3 
 
 a 
 
 1 
 < 
 
 «4H 
 
 Is 
 
 
 1 
 a: 
 
 u 
 
 a 
 
 m 
 a 
 
 € 
 
 
 ■s 
 
 a 
 
 u 
 
 
 
 43 
 
 
 liusstila sp. : 
 
 Original material . . . 
 
 Water free substance 
 Feziza auricula : 
 
 Original material .. 
 
 "Water- freesubstance 
 Agaricus campestris culi- 
 naris : a 
 
 Pervt. 
 
 8.40 
 
 10.38 
 
 92.84 
 89.54 
 
 10.58 
 21.85 
 
 Perct. 
 15.4-2 
 16.83 
 
 4.14 
 4.62 
 
 3.40 
 47.47 
 
 C3.06 
 29.8 
 
 20.93 
 23. 41 
 
 25.70 
 32.88 
 
 Per ct. 
 12.87 
 14.05 
 
 2.93 
 8.27 
 
 Per ct. 
 2.55 
 2.78 
 
 1.21 
 1.35 
 
 Pr. ct. 
 2.56 
 2.79 
 
 1.36 
 1.52 
 
 .07 
 .96 
 
 .50 
 4.80 
 
 1.19 
 1.33 
 
 .17 
 
 .22 
 
 Per ct. 
 21.41 
 23.37 
 
 34.68 
 38.69 
 
 Perct. 
 .00 
 .00 
 
 .00 
 .00 
 
 Pr. ct. 
 6.63 
 7.24 
 
 .00 
 .00 
 
 Pr. c<. 
 12.97 
 14.16 
 
 27.04 
 30.17 
 
 .54 
 7.55 
 
 .91 
 
 8.7 
 
 4.07 
 4.55 
 
 Per ct. 
 4.54 
 4.96 
 
 2.11 
 2.35 
 
 .76 
 10.67 
 
 1.08 
 10.4 
 
 7.50 
 8.39 
 
 14.60 
 18.68 
 
 Pr. ct. 
 28.07 
 30.61 
 
 20.29 
 22.64 
 
 
 
 
 
 
 
 
 Morchella esculenta : h 
 
 
 
 
 
 
 
 dl.60 
 dVo.Z 
 
 .00 .00 
 
 
 
 
 
 
 Nostoc commune jiagelli- 
 forme : 
 
 Original material 
 
 
 
 
 Porphyra laciniata : e 
 Original material 
 
 
 
 ' 
 
 
 /37. 68 
 f 48. 99 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a Reported by Dahlen, Landw. Jahrb., 4 (1874), p. 639. 
 
 b Reported by Mendell, Amer. .Tour. Physiol., 1 (1898), p. 225. 
 
 c Total nitrogen 0.49, albuminoid nitrogen 0.37, and nonalbuminoid nitrogen 0.12 per cent. 
 
 d Approximation. 
 
 e Unpublished analysis from Chem. Lab. of Connecticut (Storrs) Station. 
 
 /Determined by difference. 
 
 MISCELLANEOUS SUBSTANCES. 
 
 In addition to the above materials there are a number of others which 
 are used to a greater or less extent. The following were examined. 
 Although they did not seem important enough to demand an analysis 
 they are worthy of at least a passing notice. 
 
 The roots of an aroid, which seems to be AmorphopJiaUus rivieri, are 
 frequently observed on the vegetable stands in the Chinese quarter of 
 San Francisco. Plants grown from them at Berkeley have thus far failed 
 to produce flowers, so that their absolute identification is as yet imi^os- 
 sible. Henry,' however, refers to the use of the plant by the Chiuese, 
 but uses a character to designate it which differs from that used in San 
 Francisco. The roots, known as " kai chau u " (^|^^), are exceedingly 
 acrid, and the directions given when the samples were purchased were 
 to boil for several days before using. They are said to possess medicinal 
 or at least restorative properties, the roots being used especially by 
 persons who need a light diet. The roots sold in San Francisco, which 
 are figured in PI. Ill, fig. G, are of interest on account of the peculiar 
 compound starch grains which they contain. The roots of the closely 
 related Gonophallus Jconjalc, which are largely used by the Japanese 
 in much the same manner, have already been analyzed by Kellner,^ 
 who believed them to consist largely of starch. The later investiga- 
 
 ' Notes on the Economic Botany of China, p. 35. Shanghai, 1893. 
 2 Landw. Ver. Stat., 30 (1881), p. 42. 
 
48 
 
 f* 
 
 tions of Tsuji^ have shown that the carbohydrate material consists 
 quite largely of mannane and contains no starch. 
 
 The rhizomes of common ginger, Zingiber officinale, are largely nsed 
 by the Chinese in California and may be obtained in a fresh condition 
 throughont the year. 
 
 Among green vegetables the common purslane, Portnlaca oleracea, is 
 sometimes seen on the vegetable stands, but it is not extensively used. 
 Purslane is used to some extent as a pot herb in American households. 
 Its suitability for this purpose has been i)ointed out by Coville.^ 
 
 According to an analysis reported by Huston,^ purslane has the fol- 
 lowing percentage composition: Water, SG.oG; protein, 1.81; fat, 0.50; 
 nitrogen-free extract, 6.49; crude fiber, 2.12; ash, 2.23. It is noted that 
 the ash content is unusually high. 
 
 The young shoots of a bamboo, probably a species of Arundinaria, 
 are also sold, either pickled in brine or canned. 
 
 Of seeds, those of the common watermelon are very commonly eaten, 
 usually raw. Two varieties of sesame seed, one black, the other white, 
 are used in soup as barley is used in American households. A peculiar 
 variety of the almond and another of the chestnut are also imported 
 from China. 
 
 Among fruits, dried i^ersimmons are often met with. The persimmon 
 seems to be well adapted to this method of preservation. The seeds 
 of Areca catechu, or betel jjalm, are often sold on the street corners in 
 the Chinese quarters of San Francisco. A combination of slices of 
 this nut with lime and the leaves of a species of piper, as used by the 
 betel chewer, may also be obtained. However, they could not be 
 identified. 
 
 The Chinese drug stores also furnish some curious objects of a veg- 
 etable nature. Among these were identified, chiefly through the report 
 of Hanbury ^ on the Chinese materia medica, the fruit stalks of Hovenia 
 dulcis, the seeds of Torreya nuci/era, Quisqualis indica, Momordica 
 {Muricia) cochinchinensis, Terminalia chehula, Sterculia platani/oliaj 
 Ipomaa hederacea, Chalmougra sp., Melia sp., and the fruits of a 
 species of Gardenia. 
 
 ' Imp. Univ. Col. Agr. [Tokyo] Bui., Vol. 2 (1894), p. 103. 
 
 2 U. S. Dept. Agr., Yearbook 1895, p. 213. 
 
 3 Indiana Sta. Rept. 1897, p. 16. 
 
 ■• Scieutitic Papers, 1876, pp. 228-254, 
 
 Lb Mr '07 
 
LIBRARY OF CONGRESS