Bulletin No. 44 
 
 U. S. DEPARTMENT OF AGRICULTURE 
 
 DIVISION OF CHEMISTRY 
 
 SB 
 
 SWEET CASSAVA: 
 
 ITS CULTURE, PROPERTIES, AiND USES 
 
 SY 
 
 HARVEY W. WILEY 
 Chemist of i?he U. S. Department of Agriculture 
 
 PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE 
 
 WASHINGTON 
 
 GOVERNMENT PRINTING OFFICE 
 1894 
 
 Cop\ 
 
 I™ 
 
 -rh 
 
Book__^Jl_3\Y^ 
 
Bulletin No. 44 
 
 U. S. DEPARTMENT OF AGRICULTURE 
 
 DIVISION OF CHEMISTRY 
 
 SWEET CASSAVA: 
 
 ITS CULTURE, PROPERTIES, AND USES 
 
 BY 
 
 HARVEY W. WILEY 
 Chemist of thk U. S. Depaktmekt of Agricultuke 
 
 PUBLISHED BY AUTHORITY OF THK SECKETAKY OF AGRICULTURE 
 
 WASHII^GTOK 
 
 GOVEKNMENT PRINTING OFFICE 
 1894 
 
 T] 
 

 %^^ 
 
 N(^l^ 
 
 LETTER OF TRANSMITTAL. 
 
 U. S. Depaktmkxt of Agriculture, 
 
 Division of Chemistry, 
 IVasMnfjton, D. C, September 15, 1894. 
 
 SiH: I beg to submit for yonr insppotion and approval the manuscript of Bulletin 
 No. 44, giving a description of our investigations of the agricultural possibilities 
 of sweet cassava. 
 Respectfully, 
 
 H. W. Wiley, 
 Chief of Division. 
 Hon. Charles W. Dauney, Jr., 
 
 Act'iny Secretary of AgricuJiure. 
 
 CONTENTS. 
 
 Page. 
 
 Distribution — names — varieties 3 
 
 Cassava as an article of food 4 
 
 Analysis of cassava roots, exclusive of bark 4 
 
 Analysis of cassava flour 5 
 
 Extracts from correspondence 6-8 
 
 Analysis of root and bark, each separately 9 
 
 Comparison of cassava starch with Indian corn starch 10 
 
 Composition of the ash 12 
 
 Methods of culture 13 
 
 Conclusions 16 
 
 9 
 
SWEET CASSAVA: ITS CULTURE, PROPERTIES, AND USES. 
 
 DISTRIBUTION — NAMES — VARIETIES. 
 
 In the southern peninsula of Florida, and growing up well into the 
 frost belt, is found in many localities a cultivated plant known as 
 cassava, or sweet cassava. From a careful study of the climatic con- 
 ditions under which the plant flourishes it is safe to assume that it may 
 also be grown with success in southern Alabama, Mississippi, Louisiana, 
 and Texas. Cassava is a name which should properly apj^ly only to the 
 purified starch derived from the roots of the plant, but it has passed 
 into general use to designate the plant itself. Botanically, the plant is 
 known as Janipha inanihof, Manihot ntilissima,, Jatropha nianihot, Man- 
 ihot a'ipi,, Manihot loeJ!inf/ii, and Manihot palmata. One of its common 
 names is manioc plant. The fleshy root of this plant yields the greatest 
 portion of the daily food of the natives of many portions of tropical 
 America, and one of its forms of starch is iinj)orted largely into this 
 country as tapioca. It is a woody or shrubby plant, growing from 
 fleshy, tuberous roots, the stems being smooth, with nodules where the 
 leaves grow. 
 
 There is properly only one variety of the plant growing in Florida, 
 while that, variety which grows in the tropics contains so much hydro- 
 cyanic acid as to render it poisonous. The variety grown in the sub- 
 tropical region of Florida, however, contains only a small quantity of 
 hydrocyanic acid, and is therefore commonly known as sweet cassava. 
 Some of the growers of the plant in Florida claim that two varieties 
 grow in the State, one of which is poisonous on account of the large 
 amount of hydrocyanic acid ^yhicll it contains, and the other nonpoison- 
 ous, containing only a little hydrocyanic acid. It is quite probable, 
 however, that after the poisonous variety has grown for a long while 
 in a subtropical climate it would lose largely its i)oisonous properties. 
 The leaves of the poisonous variety in the trojDics usually have seven 
 branches palmately divided. The leaves of the sweet variety are usually 
 only five-parted. The botanists clearly recognize two distinct varieties. 
 For instance, in the "Treasury of Botany," page 718, the following 
 remarks are made: 
 
 It is quite clear that while the root of one ia bitter and a virulent poison that of 
 the other is sweet and wholesome, <iud is commonly eaten cooked as a vegetable. 
 Both (if them, especially the bitter, are most extensively cultivated over the greater 
 
 3 
 
part of tropical America aud yield aii abuudauce of wholesome and nutytious food 
 the poison of the bitter kind being got rid of during the process of preparation it 
 undergoes. The poisonous, expressed juice, if allowed to settle, deposits a large 
 quantity of starch known as Brazilian arrowroot or tapioca meal, from which the 
 tapioca of the shops is prepared by simx)ly torrefying the moist starch ujion hot 
 plates, the heat causing the starch grains to swell and burst and become agglutinated 
 together. A sauce called cassarecp, used for flavoring soups and other dishes, partic- 
 ularly the West Indian dish known as pepper pot, is also prepared from this juice by 
 concentrating and renderiug it harmless by boiling. Another of the products of 
 cassava is an intoxicating beverage called piwarrie, but the manner of preparing it 
 is not calculated to render it tempting to Europeans. It is made by the women who 
 chew cassava cakes and throw the masticated materials into a wooden bowl, where 
 it is allowed to ferment for some days, and then boiled. It is said to have an agree- 
 able taste. 
 
 CASSAVA AS AN ARTICLE OF FOOD. 
 
 The- atteutiou of the Division of Chemistry was first called to the 
 cassava plant as au article of food and a possible source of starch, in 
 1888, in a letter received from Mr. R. H. Burr, of Bartow, Fla. Mr. 
 Burr also sent a package of cassava roots. He described the plant 
 and its uses in the following words : 
 
 The roots do not last long after digging, dryiug up, or rotting. Since this variety 
 of cassava is not the bitter or poisonous kind, it is generally known in Florida as 
 the sweet cassava. The roots are fed to all kinds of stock in a fresh state, aud are 
 greatly relished. It has been sufficiently tested here to show its great value as a 
 stock food. The yield under favorable conditions is astonishing. I have recently 
 dug one plant of one year's growth which weighed .50 pounds, being at the rate of 
 more than 1,500 bushels to the acre. Eight hundred to 1,000 bushels per acre cau be 
 confidently reckoned on. 
 
 The roots received by us were long and slender and of various sizes; 
 some of them were quite 2 feet long, and weighed several pounds. The 
 analysis of the sample, exclusive of the bark, calculated to dry sub- 
 stance, is given in the following table : 
 
 Serial number 5547 
 
 Ash per cent.. 1. 94 
 
 Oil (petroleum ether extract) do.. .. 1. 27 
 
 Ether extract (resins, alkaloids, organic acids, etc.) do 74 
 
 Alcohol extract (aniids, sugars, gliicosids, etc.) do 17.43 
 
 Crude fiber do ... . 4. 03 
 
 Starch do .... 71 . 85 
 
 Albuminoids (calculated from nitrogen) do 3. 47 
 
 100. 73 
 
 In regard to the method of analysis little need be said ; it was carried 
 on in accordance with tlie well-established rules of i)lant analysis, as 
 laid down by Dragendorff. The first extraction of petroleum ether 
 gave the fat or oil alone, and the subsequent extraction Avith sulphuric 
 ether gave the resins, alkaloids, and organic acids. That portion of 
 nitrogen existing as amids was estimated in the alcoholic extract. 
 The total nitrogen was also estimated and entered as albuminoids j a 
 
small portion of the nitrogen has thus been counted twice in the total 
 results, which add up a little over 100. A characteristic feature of the 
 cassava root is shown in the large amount of substance Dresent, solu- 
 ble in alcohol. The amount of starch also compares fairly well with 
 the best varieties of potatoes. On account of the large quantity of 
 sugars present, the cassava root could be more economically used for 
 the manufacture of glucose than for starch. There is no doubt, how- 
 ever, of the fact that a tine article of starch for the laundry or for food 
 can be made from the cassava root growing in this country. 
 
 In addition to tlie fresh root above noted, two samples of the dried 
 root or cassava meal have also been examined. No. 5922 was sent to 
 us, described as pulverized manihot root or cassava flour. The root is 
 first peeled, chopped into thin slices, dried in the sun two days, and 
 pulverized. It was prepared by Prof. W. H. Kern, of Bartow, FJa. 
 1^0. 5023 was labeled pulverized cassava, with the starch, or a portion 
 of it, and glucose washed out, the remaining pulp dried in the suuj 
 prepared by Prof. Kern. 
 
 Prof. Kern sent a letter with the samples, from which the following 
 extracts are made : 
 
 Allow me to say that, owing to the prodigious yield per acre of what we here know 
 as cassava and its alleged value as a feed and food jilant and for its yield of starch and 
 glucose, it is attracting a very great deal of attention here now. The plant here grown 
 is different from the manioc root of South and Central America; our root contains 
 no poisonous elements which need to be dissipated by heat. It is customary here, 
 for many persons to make their own starch from it. The root, which must remain 
 in the ground until one is ready to use it, is dug, washed, and its two inner and 
 outer x>eelings removed; it is then grated and the pulp washed, the water poured 
 off in a vessel and allowed to stand, when the pure starch settles in the bottom. The 
 clear water is again drawn off and the starch allowed to dry. The pulp, after 
 having the stnrch washed out, may be used at once in making puddings by the 
 addition of milk, eggs, etc. This washed pulp may be sun-dried and thus kept, 
 forming valuable meal or Hour, from which nice bread may be made. Necessitated 
 as we are in south Florida to buy all our wheat flour, anything which acts as a sub- 
 stitute, either in whole or in part, is of great value to us. 
 
 The analyses of two samples of flour are given in the following 
 table : 
 
 Serial number 5922 5923 
 
 Water per cent.. 10.56 11.86 
 
 Ash do.... 1.86 1.13 
 
 Oilandfat do 1.50 .86 
 
 Resins, alkaloids, and organic acids do 64 .43 
 
 Amids, sugars, glucosids do 13.69 4.50 
 
 Dextrin, gum, etc., by difference do 2.8.5 5.63 
 
 Crudefiber do 2.96 4.15 
 
 Nitrogenous bodies do 1.31 1.31 
 
 Starch do.... 64.63 70.13 
 
 From the above analyses it is seen that the cassava can never take 
 the place of the flour made from cereals as a food material on account 
 of the small portion of nitrogenous matter which it contains. It seems 
 
to me, however, that it might very well take the place of potatoes, and 
 its value as a food should not be underestimated. 
 
 In order to get further information in regard to the growth and uses 
 of cassava in Florida, a circular was sent to as many addresses as could 
 be obtained of persons interested in its growth in that part of the State 
 suited to its culture. This circular asked for information in regard to 
 the time and method of i)lanting, method of cultivation, kind of soil 
 and fertilizers, proper time for harvesting, length of time the roots can 
 remain in the ground, the effect of frost on the plant, the yield of roots 
 per acre, their value as food, method of preparing for food and starch, 
 profit of cultivating, and the certainty of the crop. In all, about three 
 hundred answers were received to these questions. Some of these 
 answers stated that the crop was not grown in the part of the State 
 from which they were written. A few of tlie replies were unfavorable 
 in regard to the utility and value of the crop. Nearly all of them, 
 however, represented that the crop was a certain one, extremely valu- 
 able, and the roots an excellent food for man and beast. 
 
 Many of the correspondents give rather full notes in regard to the 
 value of the crop and the localities in which it grows. In Florida it 
 seems to flourish in almost every part of the State. 
 
 Mr. J. B. Pixton, of Ellerslie, Pasco County, writes: 
 
 I found it growing profusely in Taylor County, this State, some yaars since. Tbe 
 roots were thicker than a man's arm and about 4 feet long. The settlers almost live 
 " on it. ' 
 
 Taylor County is in the northwestern part of the State, bordering on 
 the Gulf. The same writer says that in his locality it did not do so well 
 and that he could make nothing of it. 
 
 Prentice Bailey, in an article in the Tropical Sun of February 9, 1893, 
 published in Juno, Dade County, says: 
 
 I consider it superior to any root crop grown in this country. It is very produc- 
 tive; it has a remarkable immunity from drought, ilood, and disease; it is easy to 
 harvest, easy of cultivation, and occupies tho ground during the whole growing 
 season to the exclusion of noxious i)lauts. The pork made from feeding it is solid 
 and delicate as chicken, and the lard is as tirni as that of corn-fed hogs. It produces 
 a good flow of rich milk and a firm, golden butter. From 1 acre of cassava enough 
 roots may be obtained to fatten 10 hogs or feed 3 milch cows during the entire year- 
 Mr. Bailey states that in his opinion it can be grown as far north as 
 Atlanta, Ga., in which opinion, however, I am unable to concur, as pre- 
 ponderating testimony goes to show that the plant will not do well in 
 localities subject to heavy frosts. 
 
 Mr. Fred W. Inman, of Winter Haven, Polk County, says: 
 
 There is no better food for stock, especially during the winter months. I have fed 
 my cows, mules, horses, hogs, and poultry upon it everj^ season for the past six 
 years. It fattens better than corn. Tliero is nothing grown in this country that 
 can compare with it. It has no enemy but fro.st. I esteem this crop one of the most 
 essential as well as profitable that the Florida farmer can grow. It furnishes- food 
 for my stock, poultry, etc., for about nine months in the year. 
 
Mr. Sherman Adams, in the Florida Agricnlturist, says: 
 
 Soil suitable for corn is appropriate for cassava. It must not, however, be wet 
 land or land subject to overflow, as that will rot the tubers. Frost, if severe, will 
 kill the plant so effectually that but a small proportion will sprout again. By sav- 
 ing the stumps when tlio roots are dug, and planting them, they will sprout and 
 grow, though the tops be killed two or three times. There are about 2,740 hills per 
 acre. On land that will not grow more than 5 bushels of corn per acre, cassava 
 will average from ^ to 5 pounds per hill, or, at a very moderate estimate, 4 to 5 
 tons per acre. 
 
 Mr. S. W. Carson, of Midland, Polk County, says: 
 
 There is but one variety of cassava, viz, the sweet kind, gfown in Florida. I have 
 never seen any sample of the bitter variety. A Spaniard who was once conversing 
 with mo on the subject of cassava gave his idea as a proper definition of cassava 
 as "the life of man." After cultivating the plant for a quarter of a century, I am 
 ready to agree with him. If judiciously used it will reduce the grain rations for 
 horses and mules at least one-half. For cows you may keep bits of it mixed in the 
 slops and other food. I have never cooked it for stock, as I believe it is best fed raw. 
 In feeding to fowls, it should be thrown into the yard in its raw state daily, but in 
 small quantities. After feeding on it for one mouth any fowl will be fat enough to 
 cook without lard or butter. Honey bees forage largely on the blooms; cattle eat 
 with relish the tender shoots. The finest fowl yard imaginable could be made in a 
 cassava patch by turning the fowls into the patch ten mouths after ]»lanting. The 
 roots would supply them with food and the tops shelter them from the sun. 
 
 1 regard the rolling pine lands, containing some willow oak, to be the best for 
 cassava, and the southern counties to be best suited to it. Let the soil be well pre- 
 pared by plowing and harrowing, rows checked about 4 feet apart, and one piece laid 
 in each hill. I think they should never be closer together than 4 feet, and 5 would 
 be better. Cassava has been known to grow for three years in this country. It will 
 continue to grow until the cold kills it; then, by breaking oft' the stems when they 
 are red, the stubble will sprout up in the spring. As to the seeds of the cassava, 
 they will ripen in about one year. If puddings, custards, etc., are desired, the roots 
 must be peeled and grated; salt, sugar, etc., may be used according to taste. The 
 Spaniards make bread of it simply by grating the root, and adding salt and a little 
 soda. Now, there is no doubt in iny mind but that 30 tons of cassava root per acre 
 can be produced. When I think of the tapioca, glucose, and starch there are in it, 
 and how abundantly it can be turned into bacon and lard, ndlk and butter, mutton 
 and beef, I feel confident that it will pay better than any other plant in the world. 
 
 Mr. Paul Dupuy, of Boardman, Marion County, says but little cassava 
 
 is grown in that locality: 
 
 Some months ago I spent fourteen months in Brazil, where the mandioc plant is 
 generally used as food by man and beast. In truth, it constitutes the bread of the 
 country, being a general article of food for all classes. It is prepared for food in 
 Brazil by grating it into a coarse pulp and pressing it to get rid of as much of the 
 juice as possible, which contains a large proportion of hydrocyanic acid. The pulp 
 is then placed in shallow copper pans and thoroughly dried over a gentle fire. In 
 this condition it resembles corn grits, and it is eaten in this shape, or it can be 
 cooked and jirepared as corn meal and other starchy products. The sediment from 
 the expressed juice, when dried, constitutes the tapioca of commerce. As a starch 
 product I do not think it can be excelled, because of its enormous product per acre. 
 
 Mr. A. Stephens Means, of Johns Pass, Hillsboro County, says : 
 
 It has great value as stock food, being very rich in starch, and most animals eat 
 it greedily. It is a nourishing food for man and can be used in a number of ways, 
 as a substitute for potatoes, or the starch may be extracted and used for puddings, 
 etc. Any soil suitable for potatoes cau be used for cassava. 
 
Mr. M. K. Lyman, of Lautana, Dade County, writes : 
 
 I have no success with cassava. I have made several efforts to raise tBe plant and 
 have failed in every case. I have splendid success ■with tanyah, or taro. It makes 
 a substitute for the potato all the year, being an evergreen, yielding edible roots the 
 year around. 
 
 Mr. William Fisher, of Clay County, says: 
 
 I doubt very much if cassava ever supplants the sweet potato as human food, for the 
 Southerners like the sweetness of the potato and it can be cooked in one-third the 
 time required for cassava. But as stock feed I believe the cassava is well worth a 
 trial. lu my judgment, the planting, cultivation, and harvesting of it would involve 
 less labor than the potato; the seed is as easily wintered; the crop can be left in the 
 ground all winter, and it yields more per acre than the potato and is apparently 
 equal to it in nutritive value. 
 
 Mr. Stephen Powers, editor of the Florida Dispatch, says: 
 The plant is indigenous to the West Indies and to Africa. In the West Indies it 
 is known as cassava, while in South America it is commonly known as mandioca, or 
 mauioca. The name yucca, which some people insist on giving it, is a clear mis- 
 nomer, since the yucca belongs to the LiUacew, or lily family, while the cassava 
 belongs to the Eiq^horhiacew family. The plant was eaten by the South American 
 and Caribbean Indians centuries before this hemisphere was discovered by Euro- 
 peans. Southey says of it : " If Ceres deserved a place in the mythology of Greece, 
 far more might the deification of that person have been expected who instructed his 
 fellows in the use of mandioca." In Brazil, in the form of coarse flour, a majority 
 of the population use it the year round in lieu of bread. A mild intoxicant is also 
 distilled from it which is the universal drink of the lower classes. The South 
 American Indians use the following formula for manufacturing cassava beer: The 
 roots are sliced, boiled till soft, and cooled. They are then chewed mouthful by mouth- 
 ful, the masticated cuds being returned to the vessel. [This process, although not so 
 stated in the description, is evidently intended to transform the starch into sugar 
 by the act'on of the saliva.] The jar is then filled with water and boiled for several 
 hours with constant stirring. The liquid is then poured off into another j ar, half 
 buried in.the dirt floor of the hut in which the family lives, closely covered, and 
 allowed to ferment two or three days. When required for use a fire is built around 
 it, and the l)everage is served steaming hot. It is not as palatable for human food 
 as sweetipotatoes, and to expect therefore to substitute cassava for either the sweet 
 or the Irish potato would be a step from a higher to a lower civilization, but as a 
 food for%live stock it has a great superiority. It is worth at least 25 per cent more 
 than sweet potatoes to produce milk or fat. With no more manure, and loss culti- 
 vation than is required for Indian corn, it will produce an amount of feed worth at 
 least four times as much for fattening animals aud incomparably more for producing 
 milk. It is easier toj)lant and cultivate than sweet potatoes, but harder to dig, so 
 it maybe called even on this score. With the same manuring it will yield 600 
 bushels per acre, while sweet potatoes will yield only 500. 
 
 Mr. J. H, Moore, of Keuka, Fla., in a letter to the same paper of 
 
 November 24, 1887, describes some of the uses of cassava. From his 
 
 letter the following extract is made: 
 
 Cut the stalks about 1 inch above the ground, just before frost; after cutting, the 
 stalks should bo left to dry in a cool place a few weeks, and then placed in a trench 
 and covered until time for planting. Some save the stalks by keeping them in a dry, 
 cool place until February and then planting. The roots should be dug as used ; they 
 will not keep in good condition out of the ground more than three or four days. It 
 is perhaps the best feed we can raise for hogs; it is also a fine feed for poultry. We 
 often bake it like sweet potatoes, aud als© slice and fry it like Irish potatoes. 
 
9 
 
 M, Sacc has addressed a letter to the National Society of Agriculture 
 of France concerning the cassava, which he calls Manihof nUlissima. 
 He is of the opinion that the poisonous varieties are different botan- 
 ically from the iHinocent. Manihot is the bread of tropical regions. The 
 innocent variety is cultivated in Bolivia, and the botanists there call it 
 Manihot dipi. The plant grows from 1 to 2 meters in height, with 
 straight and naked stalks, since they only develop leaves at their 
 extremities; the only care given to them in their cultivation is to keep 
 them free from weeds. The roots, to the number of five to nine, are of 
 the size of the closed hand. The following analysis of the roots of the 
 Manihot aipi is given: 
 
 Per cent. 
 
 Water 70.29 
 
 Starch 14.40 
 
 Sugar, salts, and malic acid 1.01 
 
 Fibrin and yellow coloring matter 08 
 
 Crude fiber 3.16 
 
 Ash 10.82. 
 
 From the above it is seen that the roots of the tropical plant are quite 
 different from those produced in our own country. In regard to the 
 distribution of the two varieties, M. Sacc makes the following observa- 
 tion : 
 
 In Cuba I have seen only the jioisonous variety. The same is true of Brazil, where 
 I have not seen the Manihot dipt except in the Swiss colony, Porto Real. As to the 
 product of the two varieties, it is the same; the stalks, which are the size of the 
 finger, are from 1 to 2 meters in height. I have not been able to analyze the leaves 
 of this interesting A^egetable, but as they are much sought after by cattle they are 
 probably very nutritious. 
 
 These interesting facts concerning the cassava plant, derived from 
 our own analyses and observations and from the correspondence noted 
 above, have led us to believe that a wider knowledge of the properties 
 of this interesting plant would prove of interest not only to the growers 
 thereof, but also to capitalists and manufacturers, who might be inter- 
 ested in it as a promising source of food and as the raw material for 
 the manufacture of starch and glucose. 
 
 A large quantity of the root, therefore, was obtained from Florida, 
 the bark separated from the root, and each subjected to analysis with 
 the following results : 
 
 
 Peeled root. 
 
 Fiber 
 
 after 
 
 removal 
 
 of starch. 
 
 Bark of root. 
 
 
 Fresh. 
 
 Per cent. 
 
 61. 30 
 
 .17 
 
 .64 
 
 30.98 
 
 88 
 
 .51 
 
 5.52 
 
 Dry. 
 Per cent. 
 
 Dry. 
 
 Fresh. 
 
 Dry. 
 
 Moisture 
 
 Percent. 
 
 Per cent. 
 
 61.30 
 
 .66 
 
 2.29 
 
 Per cent. 
 
 Ether extract 
 
 .44 
 1.66 
 
 80.06 
 2.26 
 1.31 
 
 14.27 
 
 .30 
 
 1.02 
 
 64.64 
 
 10.68 
 
 1.42 
 
 21.94 
 
 1.70 
 
 Albuminoids (mtrogeu ,\ 6.35) 
 
 5.91 
 
 Starch (diastase-extract inverted with HCI) 
 
 
 Fiber 
 
 3.83 
 
 2.02 
 
 29.90 
 
 9.89 
 
 Ash 
 
 5.23 
 
 Undetermined 
 
 77.27 
 
 
 
 
 100. 00 
 
 100. 00 
 
 100. 00 
 
 100. 00 
 
 100. 00 
 
10 
 
 With the starch in the analysis given above is reckoned also the solu- 
 ble carbohydrates, consisting almost exclusively of cane sugSr, and of 
 which, in an analysis«of another portion of the dry substance, as high 
 as 17 per cent were found. In the laboratory it is not difficult to pre- 
 pare crystallized cane sugar from the aqueous extract of the fresh pulp. 
 We have made such a preparation. The percentage of sugar iu the 
 plant, however, Istoolow to excite anyreasonable hope of the preparation 
 of this. article ou a commercial scale. The most promising way to save 
 it is by conversion into glucose, as indicated in another place. The 
 undetermined portion consists of the digestible fiber and carbohydrates 
 of the pentose series. The i^entosans in the fiber were determined by 
 the furfurol process, as modified by Krug, and the amount in the air. 
 dried material was found to be 3.92 per cent, and iu the material after 
 the removal of the starch 5.33 jier cent. 
 
 The fresh root was found to contain 38.7 per cent of dry matter, being 
 considerably more than was found in the fresh sample of the previous 
 analysis. Of this 38?7 per cent, 30.98 consisted of starch and soluble 
 carbohydrates. 
 
 Experiments were made to determine the yield of air-dry starch which 
 could be obtained from the roots by laboratory work. Two sets .of 
 experiments were made. In the first set the roots were pulped on a 
 Pellet rasp, used for i)rei>aring beet pulp for instantaneous diffusion. 
 Twelve kilos of the unpeeled root were rasped in this way and the starch 
 separated by washing through a sieve of bolting cloth. The.washings 
 and settlings were collected and dried. in. the ordinary method of starch 
 manufacture. The yield of pure starch was 3,105 grams, equivalent to 
 25.9 per cent of the total weight of the root. The starch was almost 
 absolutely pure, containing only a trace of nitrogenous. matter. In the 
 sec.ond experiment 10 kilos of the root were ground in a pulping machine, 
 used for preparing green fodder for analysis. The pulp was much less 
 fine than that produced by the Pellet rasp. Trea.ted in the. same way, 
 the yield of air-dry starch was 2,360 grams, or 23.6 per cent. One of the 
 striking points in connection with the work is that the residue from the 
 starch, which consisted largely of fiber, as will be seen by reference to 
 the above analysis, contained still a large percentage of starch, show- 
 ing that by the process employed the whole of the starch was not 
 secured from the pulp. The diameter of the starch granules is a little 
 over 0,01 mm,, being about seven times smaller than the average of 
 potato starch. 
 
 COMPAKISON OF CASSAVA STARCH AND INDIAN CORN STARCH, 
 
 The illustrations given are from photomicrographs of the starch 
 granules, which, for purposes of comparison, are accompanied by illus- 
 trations of the finest pre])ared Indian-corn starch. 
 
 Fig. 1, Plate i, shows the granules of cassava starch in plain light, 
 magnified 150 diameters. 
 
11 
 
 Fig. 2, Plate i, shows Indian corn starch in siniihir conditions. The 
 average size of the jiarticles is the same in both instances, viz, 0.012 mm- 
 The cassava particles are more irregnlar in size, and are, moreover, to 
 be distingnished from the maize particles by greater evenness of ont- 
 liue. The cassava, however, more nearly resembles the maize starch 
 than it does any other well-known variety, and by reason of this resem- 
 blance conld easily take the i^lace of maize starch in the kitchen and 
 laundry. 
 
 Fig. 1, Plate ii, shows cassava starch magnitied 150 diameters and 
 viewed by polarized light. Only the larger particles are well in focus, 
 nevertheless the cross can be seen on the smaller by close inspection. 
 
 Fig. 2, Plate ii, shows maize starch in similar conditions. While it 
 would be difficult in all cases to distinguish these two starches by the 
 microscope, yet there are some points of difference, as have already 
 been noted, by means of which the expert may usually be successful in 
 the discrimination. 
 
 The cassava which grows in tropical regions contains a notable per- 
 centage of hydrocyanic acid, so great, in fact, that it can not be used 
 directly as a food. The so-called poisonous cassava is boiled, to expel 
 the hydrocyanic acid before being used for feeding purposes. A care- 
 ful determination was made of the hydrocyanic acid in the fresh root 
 and the amount was found to be in considerable quantity, but not suf- 
 liciently large to be alarming. Nevertheless, any possible danger could 
 be avoided before using the material as a food by subjecting it to a 
 sufficient heat to expel the hydrocyanic acid. The hydrocyanic acid 
 seems to be distributed throughout the pulp, and particularly in the 
 juices which can be expressed from the pulp. jSo injurious effect from 
 the hydrocyanic acid has ever been observed in the case of animals fed 
 on cassava in Florida. 
 
 The bark of the root was also subjected to analysis, as will be seen 
 by reference to the following table. It contained no starch, the unde- 
 termined matter being chieily digestible fiber and pentosans. 
 
 The mineral matters extracted from the soil are distributed as indi- 
 cated in the table. The amount of ash in the root itself is quite low, 
 showing that the cassava plant does not require a soil very rich in 
 mineral constituents. The amount of mineral matter taken from the 
 soil by 100 kilos of the fresh root is approximately only half a kilo. 
 The albuminous matters are also present in small quantities, being 
 only slightly larger in weight than the ash itself. The i)lant, therefore, 
 is one which seems particularly suited to feed almost exclusively from 
 the air and water, and hence is one which could be recommended on 
 the sandy soils of Florida as a crop which would require the minimum 
 of fertilization. 
 
12 
 
 COMPOSITION OF THE ASH. 
 
 The ash of the peeled, root aud the bark of the root was subjected to 
 analysis, with the following results : 
 
 Analysis of the ash of cassava root. 
 
 Constituents. 
 
 Carbon 
 
 Silica (Sdluble in solution of Na5C03) . 
 Silica (insoluble in solution ot Na2C03) 
 
 Ferric oxiil ( lejUj) 
 
 Calcium oxid (CaO) 
 
 Magnesium oxid (MgO) 
 
 Sodium oxid (Na.,( )) 
 
 Potassium oxid (K.;0) 
 
 Phosphoric acid ( PaOs) 
 
 Sulphuric acid (SO3) 
 
 Carbonic acid (CO2) 
 
 Chlorin (CI) 
 
 Total 
 
 Oxygen equivalent to chlorin 
 
 Difference 
 
 Peeled root. 
 
 A. 
 
 Per et. 
 
 0.30 
 
 0.97 
 
 7. 15 
 
 0.66 
 
 10.63 
 
 7. 3C 
 
 1.12 
 
 41.72 
 
 15.58 
 
 3.67 
 
 9.15 
 
 2.76 
 
 B. Mean. 
 
 Per ct. 
 
 0.31 
 
 91 
 
 7. 1.") 
 
 0.66 
 
 10.64 
 
 7.35 
 
 1.28 
 
 41.54 
 
 15.59 
 
 3.80 
 
 9.12 
 
 2.75 
 
 101.07 
 0.62 
 
 101.10 
 0. 62 
 
 100. 45 
 
 100. 48 
 
 Per ct. 
 
 0.31 
 
 0.94 
 
 7.15 
 
 0.66 
 
 10.04 
 
 7.35 
 
 1.20 
 
 41.63 
 
 15.58 
 
 3. 73 
 
 9.14 
 
 2.75 
 
 101. 08 
 0.62 
 
 100.46 
 
 Bark of root. 
 
 A. 
 
 Per ct. 
 
 0.79 
 
 10. .53 
 
 52. 99 
 
 2.46 
 
 6.58 
 
 3.31 
 
 0.84 
 
 14.73 
 
 2.44 
 
 1.71 
 
 2.53 
 
 1.41 
 
 100. 32 
 0.31 
 
 100. 01 
 
 B. Mean. 
 
 Per ct. 
 
 0.77 
 
 11.30 
 
 52. 16 
 
 2.44 
 
 6.65 
 
 3. 33 
 
 1.05 
 
 14.68 
 
 2.46 
 
 1.71 
 
 2.50 
 
 1.42 
 
 100. 53 
 0.31 
 
 100.22 
 
 Per ct. 
 
 0.78 
 
 10.94 
 
 52.58 
 
 2.45 
 
 6.62 
 
 3. 32 
 
 0.95 
 
 14.70 
 
 2.45 
 
 1.71 
 
 2.51 
 
 1.41 
 
 100.42 
 0.31 
 
 100.11 
 
 From the above numbers it is seen that the ash of the peeled root is 
 especially rich in potash, almost one-half of the total weight being com- 
 posed of this substance. The potash is combined chiefly with carbonic 
 and phosphoric acids. In the ash of the bark, as might be expected, 
 silica IS the predominant element, more than half the total weight con- 
 sisting of this substance. 
 
 Assuming a yield of 5 tons of roots per acre, the weights of the 
 important fertilizing materials removed by such a crop can be readily 
 calculated from the data given. 
 
 Since the bark forms approximately 2.2 ])er cent of the entire root, 
 the total crop would be made up of the following amounts of bark and 
 peeled root which would contain the amounts of mineral matter given 
 below : 
 
 Peeled root . . 
 Bark of root . 
 
 Pounds. 
 
 9,780 
 220 
 
 Pounds 
 of ash. 
 
 49.88 
 4.44 
 
 10, 000 
 
 54.32 
 
 The more important mineral matters contained therein are: 
 
 Lime (CaO) 
 
 Magnesia (MgO) 
 
 Pota.sh (KjO) 
 
 Phosphoric acid (P2O5) 
 
 Re.sidue 
 
 Ash from 
 
 l)eeled root 
 
 (49.88 
 
 pounds). 
 
 Pounds. 
 5.31 
 3.67 
 20.77 
 
 7.77 
 
 12.56 
 
 Ash from 
 
 bark (4.44 
 
 pounds). 
 
 Pounds. 
 
 0.29 
 
 .15 
 
 .65 
 
 .11 
 
 3.24 
 
 Total ash 
 
 from 5 tons 
 
 (54.32 
 
 pounds). 
 
 Pounds. 
 5.60 
 
 3.82 
 
 21. 42 
 
 7.88 
 
 15.60 
 
13 
 
 The less valuable mineral plant foods, that is, those that are of so lit. 
 tie note as to recinire no conservation or addition, amount to 15.60 
 pounds per acre and the more valuable to 32.72 pounds iter acre. 
 
 Quite a number of preparations was made from the starch of tlie root, 
 and amony- them may be mentioned : First, tapioca; the first portions 
 of starch washed out, especially, produce an excellent article of tapioca 
 when treated in the usual way. Second, glucose; both the fresh root 
 and the extracted root yield full theoretical amounts of glucose, and 
 samples of this article were made by the conversion of the starch both 
 by sulphuric acid and diastase. The samples of glucose made from the 
 starch were exceptionally good, especially when diastase was used, 
 the glucose in this case containing large quantities of maltose. Com- 
 mercially it would be more profitable to make the glucose directly from 
 the fresh root, in which case the considerable percentage of cane sugar 
 contained by it would be saved, whereas if glucose be made from the 
 starch the cane sugar is previously washed out. On account of the 
 presence of the bark, however, the glucose nmde from the whole root 
 is not so tine in quality as that made from the pure starch. Third, 
 alcohol; the glucose on fermentation affords the usual quantity of 
 alcohol. Fourth, cane sugar; a beautiful preparation of cane sugar 
 was made from the water used in washing out the starch. The amount 
 of cane sugar, however, is not large enough to warrant its extraction 
 on a commercial scale from the waters used for washing. It is, how- 
 ever, present in sufticient quantity to indi- 
 cate that in making glucose it is better to 
 use the whole root as indicated above. 
 
 The general result of the analytical 
 work is such as to establish the fact that 
 the cassava is a plant of high economic 
 value and worthy of the attention of those 
 interested in the carbohydrate products 
 of the country. 
 
 METHODS OF CULTURE. 
 
 Cassava has been grown for one year 
 on the Department Experiment Station 
 at Runnymede (post-office, Narcoossee), 
 Osceola County, Fla. The crop was 
 grown as food for stock. The field in 
 which the crop was grown is high pine 
 sand, with almost no other ingredient. 
 The illustration of the cassava given here- 
 with is from a photograph of a plant 
 taken from a field near the station. The fig. i. 
 
 soil on which it was grown was apparently pure sand. It represents the 
 larger plants in the field, but not by any means the largest. The illus- 
 tration (fig. 1) shows in a striking manner the stem and root develop- 
 
14 
 
 ment. This plant, of wliich the photograph is given, was 5 feet high. 
 The roots in the soil occupy a more nearly horizontal f>osition than is 
 shown in the figure. The thickened part of the stem to which the roots 
 are attached, represents the cutting from wliich the plant grew. 
 
 Attempts were also made to grow the cassava in a piece of very wet 
 muck land on the station in which sugar cane would not grow to any 
 advantage. An immense development of tops was secured, some of the 
 plants reaching a height of 10 feet and resembling young trees. The 
 root development was fair, but not commensurately increased with the 
 top growth. Some of the stems were easily 2 inches in diameter. On 
 well-drained muck land I think the crop would be large and profitable. 
 
 In sand land the planting should be preceded by the removal of 
 stumps, sprouts, etc., and the soil given a thorough plowing. It is 
 advisable to spread about 300 pounds of fine raw Florida phosphate 
 floats or about 150 pounds of superphosphate containing 12 per cent 
 available acid, to the aci^e. This may be applied as a top-dressing and 
 thoroughly worked into the soil by a deei3-running cultivator. The 
 rows should be marked out in furrows 3 to 4 inches deep and from 3J 
 to 4 feet apart. To get a good stand about double the number of cut- 
 tings required to produce 2,500 hills per acre should be planted. The 
 excess of plants can be removed by a hoe as soon as vigorous growth is 
 assured, leaving one hill each 3 or 4 feet. About 150 pounds of kainit 
 per acre should be dropped in the hills before planting, together with 
 an equal amount of cotton-seed meal, or half that amount of Chile salt- 
 peter (nitrate of soda). 
 
 The cultivation should be such as to keep the field free of all weeds 
 and the surface of the soil well stirred. While the plants are young 
 deep cultivation is not objectionable, but as soon as the root system 
 begins to develop, flat superficial culture must be practiced; not to 
 exceed 2 inches in depth. Some cultivators draw the soil to the plant 
 during cultivation so as to form a ridge at the timeof laying by. Where 
 nitrnte of soda has been used an additional 50 or 75 pounds per acre 
 should be sown broadcast Just before the final cultivation. The above 
 method is the one which should be followed for the poorest kind of sand 
 soils, where a maxinuim crop is desired. For muck soils, the cotton- 
 seed meal and nitrate of soda should be omitted and about 500 pounds 
 of Florjda phosphate floats used per acre. If sand soils are covered Avith 
 a good layer of muck before the plowing the nitrogenous fertilizers 
 may also be omitted or reduced in (juantity. ' 
 
 In ordiiuiry seasons with the treatment outlined above, a crop of 
 from 4 to 7 tons i)er acre will be secured. On sand soils containing a 
 little organic matter ai)proaching the hammock variety, a fair yield of 
 from 2 to 4 tons per acre will be secured by good cultivation without 
 fertilizing. 
 
 For seed, the stems of the unfrosted plants are cut into pieces about 
 6 inches in length, care being taken that each i>iece has two or more 
 eyes. In planting, these jjieces may be laid directly down in the fur- 
 
BuL. 44, Div. Chemistry. U. S. Dept. Agriculture. 
 
 Plate I 
 
 -up*^ (5 
 
 Cassava Starch x iso. 
 
 PLAIN ILLUMINATION. 
 
 Fid 2 
 
 
 <f 
 
 
 
 ^^% 
 y%/ 
 
 Q» 
 
 © 
 
 <^|^ 
 
 ft © 
 
 © 
 
 
 O®^e0 
 
 SO, 
 
 e<^, 
 
 Corn Starch x iso. 
 
 PLAIN ILLUMINATION. 
 
 A Hoen &Co...Lithocaustic 
 
BUL 44, Div. Chemistry, U. S. Dept. Agriculture. 
 
 Plate 2. 
 
 Fid 1 
 
 Cassava Starch x iso. 
 
 POLARIZED LIQHT. 
 
 Fig 2 
 
 «*■» 
 
 
 Corn Starch x iso. 
 
 POLARIZED LIGHT. 
 
 A Hoen &Co..Lithocaustic 
 
1 
 
15 
 
 rows and covered, but tlie general practice is to place them obliquely 
 in the furrows so that one end may not be covered. In case of a threat- 
 ening frost before a field is ready for planting the unfrosted tops may 
 be cut, thrown into heaps, and protected with leaves or trash from the 
 action of the frost. They should, however, be embedded in moderately 
 moist earth if they are to be kept before planting for any length of 
 time. In case of frost before the seed is saved the stumps, i. e., the 
 points of union of the top with tlie root, will usually be found unin- 
 jured, and these may be cut away and planted instead of the cuttings 
 just described. The larger parts of the stems immediately above the 
 ground make the best seed. 
 
 The roots should be left in the ground until they are needed for use, 
 whether for food, for starch, or for glucose. The croji can be harvested 
 at any time during the year, but the best season is from October to 
 May. The roots should not be allowed to grow more than two seasons, 
 and for most purposes it is believed that an annual harvest will prove 
 the more i^rofitable. 
 
 As is the case with all new and promising plants, the most extrava- 
 gant statements have been made in regard to the amount of cassava 
 which can be produced per acre. In many of the returns received 
 from our correspondents in Florida, statements were made in regard to 
 the yield which were entirely beyond the bounds of reason. These 
 extravagant statements, of course, did not proceed from any desire on 
 the part of correspondents to misstate the facts, but on account of 
 their misapprehension of them. Statements of yield are made as a rule 
 not upon accurately measured and weighed products, but upon a mere 
 glance over a field or the taking of a few hills. It is easy, therefore, for 
 the most honest and upright correspondent to fall into gross error in 
 regard to the amount which will be furnished by an acre. In my own 
 observation of small areas and from the accredited statements of those 
 authorities which seem to merit the highest consideration, I am con- 
 vinced that on the ordinary pine land of Florida, with proper prepara- 
 tion and cultivation and appropriate fertilization, a yield of from I to 7 
 or perhaps 8 tons per acre may be reasonably expected. It iadifticult 
 to see, however, how it is possible for such yields as have been reported, 
 viz, 40, 50, and even 60 tons per acre, to be gathered. In exceptional 
 conditions, as in the case with all crops, exceptional yields may be 
 obtained, but these must not be considered in the practical study of 
 the problem of profitable production. 
 
 The profit which the farmer may make from growing this crop, and 
 the manufacturer from using it, should, in my opinion, be based upon 
 a yield of 4 or 5 tons per acre. If it be desired to make starch from the 
 plant, we may suppose as a minimum rate of yield that 20 per cent of 
 the weight of the fresh root may be obtained as merchantable starch of 
 a high grade. On a yield of 4 tons per acre this would amount to 
 eight-tenths of a ton, or 1,600 pounds. Compare this with the weight 
 of starch obtained from Indian corn producing 40 bushels per acre. 
 
16 
 
 The yield of merchantable starch of a high grade may be placed at 35 
 pounds per bushel, which, for 40 bushels would amount to 1,400 pounds. 
 It is thus seen that the rate of yield per acre in the matter of starch 
 from cassava would be fully equal if not superior to that from Indian 
 corn. 
 
 If the matter of the manufacture of glucose be considered the esti- 
 mate is even more favorable. Our experiments have shown that after 
 the removal of the bark the whole root may be rasped and treated 
 directly for the manufacture of glucose, either by inversion with dias- 
 tase or by treating with dilute sulphuric acid. In the latter case not 
 only were the starch and sugar present in the root obtained as glucose, 
 but also a considerable quantity of the digestible liber. It is not an 
 extravagant statement, therefore, to suppose that fully 30 j)er cent, on 
 the weight of the fresh root, could be obtained as commercial glucose. 
 This would give a yield per acre of 1.2 tons, or 2,400 pounds. These 
 statements are made, of course, subject to the jiractical determinations 
 of the manufacturer of glucose and starch from this ]>laut. Attempts 
 have already been made in the manufacture of starch, but of course 
 the full development of this industry must await the investment 
 of capital and the necessary adjustment of new machinery to new 
 processes. 
 
 The object of the present bulletin is simply to point out the possi- 
 bilities of the culture of cassava, not only for the farmer of Florida but 
 also for the manufacturer and merchant. 
 
 In its preparation Mr. E. E. Ewell has conducted the chemical and 
 ^preparatory work and Mr. G. L. Spencer has taken the i)hotographs. 
 
 CONCLUSIONS. 
 
 (1) Cassava can be cultivated with safety and profit in the greater 
 part of the j)eniiisula of Florida, and probably also in southern Ala- 
 bama, Mississippi, Louisiana, and Texas. 
 
 (2) It will yield with fair treatment on the sand soils from 4 to 5 
 tons i^er acre. 
 
 (3) It will give, when properly manufactured, from 20 to 25 per cent 
 of the weight of the fresh root in starch of high grade. 
 
 (4) The starch is naturally in a pure state and no chemicals of any 
 kind are necessary in its manufacture. 
 
 (5) The starch resembles in its j)hysical properties the starch of 
 maize and can be used as a substitute therefor in all cases. 
 
 (6) An excellent article of tapioca can be prepared from the starch 
 of the cassava plant. 
 
 (7) Glucose can be prepared directly from the starch or more profit- 
 ably from the pulp of the peeled root. 
 
 (8) The plant furnishes an excellent human and cattle food, defi- 
 cient, however, in nitrogen. It would make a well-balanced ration for 
 cattle when mixed with one-fourth its weight of cotton seed oil cake. 
 
 m