■■■■:■■■■■■■■■■■■■■■■■ m wm 1 n n n H I BKfl ■H IfilS 1 HI UHnM till ititfii illlll HP IRllfttfr n ■■I '■'■■ 111 -■' Ill 111 ■■■ :w,.:i.-,;;-,; .: L IHWHiiii wwmm IffiffliliMIS Us ffi This book must not be taken from the Library building. Form No. 471 WZMZ5 «&&* ,r*u * Digitized by the Internet Archive in 2012 with funding from University of North Carolina at Chapel Hill http://archive.org/details/cottonplantitshiuni] Q>(o33 • •» 54th Congress, > HOUSE OF REPEESE 2d Session. ) \ No. 267. Bulletin No. 33. U. S. DEPARTMENT OF AGRICULTURE. OFFICE OF EXPERIMENT STATIONS. It THE COTTON PLANT: ITS HISTORY, BOTANY. CHEMISTRY, CULTURE, ENEMIES, AND USES. Prepared under the supervision of A. C. True, Ph. D., Director of the Office of Experiment Stations. WITH AN INTRODUCTION. BY OHAKLES ¥. DABNEY, Jr., Ph. D., ASSISTANT SECRETARY OF AGtRICULTUR-l WASHINGTOK: GOVERNMENT PRINTING OFFICE. 1896. V. v 4 CONTENTS. Chemistry op cotton — Continued. Seed t Fertilizing constituents 94 Proximate constituents 95 Miscellaneous chemical studies 95 Cottou-seed products 96 Cottou-seed hulls 97 Cotton-seed hran and cotton-seed feed 98 Cotton-seed kernels 99 Cotton-seed cake 100 Cotton-seed meal 102 Cotton-seed oil 102 Tables of analyses 107 Bibliography 141 Climatology and soils. By Milton Whitney 143 Introduction 143 Climate 143 Soil 153 Chemical properties of soils 153 Physical structure of soils 157 Typical soils of the cotton belt 162 The manuring of cotton. By H. C. White, Ph. D 169 Historical 169 Scientific experiments bearing upon the manuring of cotton 179 Yield and profit from the use of fertilizers on cotton as compared with yield and profit from unfertilized soil 180 Comparative values of commercial fertilizers and home manures 181 Kind of fertilizer (chemical manure) required by or best suited to cotton 183 The amount of fertilizer per acre giving best results 188 Best mode of application of fertilizers to cotton 189 Best time of application of fertilizers to cotton 189 Miscellaneous experiments 190 General couclusions 191 Methods of manuring cotton at present in general use in the United States . 194 Manuring of cotton in other couutries 195 Bibliography 196 Cultivated varieties of cotton. By S. M. Tracy, M. S 197 Introduction 197 American varieties 198 Foreign varieties 210 Origination of varieties 211 Plant selection 211 Saving seed from early maturing bolls 212 Cross fertilization 212 Change of name 214 Improvement and deterioration of varieties 214 Classification of varieties 215 Relative values of varieties 218 Bibliography 224 Culture of cotton. By Harry Hammond 225 Geography of the cotton belt 225 Centers of cotton production 226 The pine levels 226 The pine-hills region 233 CONTENTS. 5 Culture of cotton— Continued. Page. Metamorphic or Piedmont region 237 Sand-hills region 240 The prairie region 241 Oak and hickory region 248 Bluff and brown loam table-lands 251 The alluvial region 252 Red-loam lands 257 The valley region 257 The alpine regions 258 General observations on cotton culture 258 Drainage 259 Inelosures 259 Subsoiling 260 Rotation 260 Planting and cultivation 261 Period of growth 262 Shedding of forms, blooms, and bolls 263 Picking 264 Cost of cotton production 266 Cotton production in different States 269 Experiments in cotton culture by the experiment stations 271 Diseases of cotton. By George F. Atkinson, M. S 279 General nature of cotton diseases 279 Mosaic disease, or yellow leaf blight 279 Red leaf blight... 284 Shedding of bolls -. 285 Angular leaf spot 286 Frenching 287 Sore shin ; damping off; seedling rot 292 Anthracnose 293 Root rot of cotton (Ozonium) 300 Artificial cultures of Ozonium 305 Cotton-leaf blight 308 Areolate mildew of cotton 309 Cotton-boll rot 310 Root galls of cotton 311 Development and metamorphoses 314 The insects which affect the cotton plant in the United States. By L. O. Howard, Ph. D 317 The cotton worm, or cotton caterpillar 320 General appearance, habits, and life history 320 Parasites and natural enemies 324 Remedies 325 The cotton boll worm 328 General appearance, habits, and life history 328 Natural enemies 331 Remedies 331 The Mexican cotton-boll weevil 335 General appearance and method of work 335 Distribution 335 Natural history and habits 336 Popular names 339 Parasites and natural enemies 339 Remedies 339 Summary of remedies 342 b CONTENTS. The insects which affect the cotton plant, etc. — Continued. Page. Other cotton insects 343 Cutworms 343 Plant lice 344 Leaf-feeding caterpillars 345 Other insects which damage the leaves 346 Insects damaging the stalk 347 Insects injuring the boll 348 The handling and uses of cotton. By Harry Hammond 351 Stalks 351 Seed cotton 352 Storage 352 Ginning 354 Baling 360 Manufacture and uses of cotton by-products 365 Oil 365 Oil mills 366 Oil-mill products 372 Present condition and outlook of the cotton-oil industry 373 Cost and profit of the cotton-oil industry 374 Fertilizing value of tbe seed and its products 376 Feeding value of cotton seed and its products 378 Conclusions 380 Lint 381 Marketing 381 Cost of transportation 383 Distribution 384 The feeding value of cotton-seed products. By B. W. Kilgore 385 Composition of cotton-seed products 386 Digestibility of cotton products 386 Does cotton-seed meal affect the digestibility of carbonaceous foods?. 387 Feeding cotton-seed products for beef production 388 English experiments ^ 388 American experiments 392 European experiments 402 Effect of cotton-seed products on beef fats 403 Feeding cotton-seed products to sheep 403 English experiments 403 American experiments 406 European experiments 408 Feeding cotton-seed products for pork production 408 Feeding cotton-seed products to calves 410 Feeding cotton-seed products to horses and mules 411 Cotton-seed products for milk and butter production". 411 American experiments 411 European experiments 414 Effect of cotton-seed products on the quality of butter 415 Effect of cotton-seed products on churnability 419 Effect of cotton-seed products on health of animals 419 Eeferences to additional articles on feeding cotton products 421 Supplemental bibliography of cotton 423 ILLUSTRATIONS. PLATES. Page. Pl. I. A cotton field Frontispiece. II. Chart showing production and consumption of cotton, 1789-1894 42 III. Map of typical soil areas of the cotton belt of the United States 226 IV. Transformations of cotton hollworm 328 FIGURES. Fig. 1. Sea Island cotton 69 2. Upland cotton 72 3. Cotton fibers 80 4. Mosaic disease 282 5. Anthracnose 294 6. Root rot 301 7. Areolate mildew 309 8. Root galls 312 9. Egg of cotton- worm moth 321 10. Cotton caterpillar 321 11. Cotton-worm moth 322 12. Proboscis of cotton- worm moth 322 13. Cotton- worm egg parasite 323 14. Chalcia flavipes, an important parasite of the cotton caterpillar 324 15. Skin of cotton caterpillar attached to underside of cotton leaf by silk spun about the pupaj of Euplectrus comstockii 324 16. Pimpla inquisitor, one of the principal parasites of the cotton cater- pillar 325 17. Diagram of cotton field showing location of trap corn 334 18. The cotton-boll weevil 335 19. Map showing distribution of the Mexican cotton-boll weevil in 1895. . 336 20. Cotton-boll weevil; newly hatched larva in young square, etc 337 21. Mature boll cut open, showing full-grown larva, and mature boll not cut, showing feeding punctures and oviposition marks 338 22. Late fall boll, showing how beetles hide between boll and involucre.. 338 23. Feltia annexa 344 24. Feltia malefida 344 25. Pyrausfa rantalis 346 26. Schistocerca americana 347 27. Cotton-stalk borer 347 28. Homalodisca coagulata 348 29. The red bug, or cotton stainer 349 30. Appearance of different kinds of cotton bales on arrival at Trieste, Austria-Hungary 361 31. The Bessonette cylindrical cotton bale 363 32. Diagram of an oil mill 369 LETTER OF TRANSMITTAL. United States Department of Agriculture, Office of Experiment Stations, Washington, D. C, June 15, 1896. Sir: I have the honor to transmit herewith a bulletin on the cotton plant, which includes summaries of information on different topics relating to this plant considered in their agricultural bearings. The effort has been made to present such facts as would be useful to the students of our agriculture, to the investigators at the experiment sta- tions, and to that increasing body of intelligent agriculturists who are interested in thoroughly acquainting themselves with the past and present condition of our agricultural industries, with a view to discov- ering means for their improvement. Xo attempt has been made to dis- cuss the problems of cotton manufacture, the purpose being to confine the bulletin strictly within agricultural lines. The introduction to the bulletin has been written by C. W. Dab- ney, jr., Ph. D., Assistant Secretary of Agriculture, whose lifelong acquaintance with the practical problems of Southern agriculture has been supplemented by special studies of scientific questions relating to the crops of that region, particularly in connection with his duties as director of the North Carolina and Tennessee agricultural experi- ment stations. The chapter on the history and general statistics of cotton as an agricultural plant has been prepared by Mr. ft. B. Handy, of this office, who has, as far as practicable, examined the original sources of informa- tion and carefully collated the literature of this subject. Many details of the earlier history of cotton are obscure, and their interpretation will always remain largely a matter of individual opinion. The article herewith is, however, a careful and independent review of the evidence available, and the numerous references to the authorities will enable the student to examine the matter for himself should he care to pursue it further. The chapter on the botany of cotton, by Walter H. Evans, Ph. D., of this office, has involved a very considerable amount of research, the results of which have been very largely of a negative character. An examina- tion of the widely scattered literature of the histology and physiology of agricultural plants has revealed surprisingly few investigations on the cotton plant. The systematic botany of this plant is also in a very 9 10 LETTER OF TRANSMITTAL. unsatisfactory state, and it is quite difficult to make definite statements which will not be subject to more or less serious criticism. For the pur- pose of this bulletin it was deemed best to make a concise and orderly statement of the facts as they appeared to the author after a careful review of available literature without entering into discussion of dis- puted points. The chapter on the chemistry of cotton, by Mr. J. B. McBryde, chemist of the Tennessee Agricultural Experiment Station, and Mr. W. H. Beal, of this office, includes summarized statements of the results of chemical investigations of the cotton plant and detailed tables of analyses, together with a bibliography of the subject. In the chapter on climatology and soils, by Prof. Milton Whitney, chief of the Division of Soils of this Department, general considerations relating to the climate and soils of the cotton belt of this country have been briefly stated, the results of soil investigations as they affect the problems of cotton culture have been explained, and some typical soils of different cotton regions have been concisely described. A short compilation of the results of chemical analyses of soils in the cotton States, selected from the large number of analyses given by Prof. E. W. Hilgard in the Tenth Census of the United States, has been added, with a view to indicating in a general way what chemical analysis has shown regarding the soils of this region. The chapter on the manuring of cotton, by IT. C. White, Ph. D., president and professor of chemistry of the Georgia State College of Agriculture and Mechanic Arts and vice-director and chemist of the Georgia Agricultural Experiment Station, contains a brief history of the use of fertilizers for cotton and a summary of the results of experience and experiment in manuring this crop. The chapter on cultivated varieties of cotton, by S. M. Tracy, M. S., director and botanistof the Mississippi Agricultural Experiment Station, includes brief descriptions of the principal varieties cultivated in the United States, with a discussion of the ways of determining their rela- tive importance and of the methods used in the origination and improve- ment of varieties. The chapter on the culture of cotton was prepared by Mr. Harry Hammond, of South Carolina, author of the article on Cotton Produc- tion in South Carolina in the Tenth Census, and a cotton planter of long experience. For the purposes of this article the cotton belt has been divided into a number of regions characterized by more or less uniform conditions of soil and climate, and the general conditions of cotton culture in each region have been concisely described. The gen- eral results of experience in the management of this crop are also stated and the cost of production together with other economic factors affecting the great industry of cotton raising have been briefly con- sidered. It is believed that this article will afford the reader a com- prehensive view of the present status of cotton culture in this country. LETTER OF TRANSMITTAL. 11 A brief summary is appended of the relatively few experiments on cotton culture thus far reported by the agricultural experiment stations, prepared by Mr. J. F. Duggar during his connection with this office. The chapter on diseases of cotton was prepared by George F. Atkin- son, M. S., professor of botany in Cornell University, who, during his connection with the Alabama Agricultural Experiment Station, con- ducted the most extensive investigation of these diseases thus far attempted. The summarized statements contained in this article are intended for the general reader rather than the specialist. The chapter on insects which affect the cotton plant in the United States, by L. O. Howard, Ph. D., Entomologist of this Department, in- cludes a summary of the principal results of the extended investigations of the Division of Entomology and of the United States Entomological Commission, intended for the use of the general reader. Of special interest is the account of the investigations on the Mexican cotton- boll weevil, recently conducted under Professor Howard's supervision. The chapter on the handling and uses of cotton, by Mr. Harry Ham- mond, treats of the storage, ginning, and baling of cotton, the manu- facture and uses of cotton-seed meal and oil and other by-products, and the marketing of the lint. Only such information is given in this article as was deemed to belong to the agricultural side of the questions relating to the handling and use of the cotton crop. The chapter on the feeding value of cotton-seed products, by Mr. B. W. Kilgore, assistant chemist of the North Carolina Agricultural Exper- iment Station, is a summary of the results of the investigations on this subject carried on at the agricultural experiment stations and kindred institutions in this country and abroad. Throughout the bulletin careful and somewhat complete references to the sources of information have been made in footnotes, and a list of works which are not thus referred to, but which may be of interest to students of the cotton plant, is given at the end of the bulletin. The labor involved in the final arrangement of material and the preparation of the bulletin for the press has largely devolved on Mr. W. H. Beal, of this office, and a large amount of work in the collection of literature and in the working up of details has also been performed by Mr. R. B. Handy, of this office. The preparation of this somewhat comprehensive bulletin on the cotton plant has involved a very large amount of labor in the colla- tion of materials, in the selection and arrangement of the information to be given under each head, and in the securing of accuracy and rea- sonable uniformity in details. On many topics long search has revealed a surprising paucity of reliable information. It is evident that thus far very few careful investigations of the cotton plant have been made. A great field of research remains open to our agricultural experiment stations, on which they have hardly begun to enter. When we con- sider how vast are the interests involved in the cotton industry, we 12 LETTER OF TRANSMITTAL. realize the total inadequacy of tlie efforts thus far put forth to solve the perplexing problems confronting the cotton planter. This bulletin will have served an important purpose if it calls attention to the need of more thorough investigation of these problems and stimulates useful inquiries in this direction. The conditions under which the workers in our experiment stations are laboring largely prevent their making comprehensive surveys of the work already done in various lines before attempting further investigations. They are compelled too often to attack the problem which seems most immediately pressing without such preliminary study of its relations to other problems as would make their work most effective and success most probable. This bul- letin has been prepared primarily to relieve this difficulty as regards the cotton plant. The conditions of our agriculture have hitherto induced superficial methods of culture and handling of our staple crops. The need of greater attention to finer distinctions and nicer economies is beginning to be at least dimly discerned by the masses of our agricultural popu- lation. As the real situation becomes clearer there will undoubtedly be au increasing demand for such information regarding these crops as can be obtained only as the result of elaborate and far-reaching inves- tigations at our agricultural experiment stations. It is our present duty to pave the way for the successful conduct of such investigations by reviewing the past and carefully noting the conditions of the pres- ent, and to supply the materials out of which the master workmen of the future may construct lasting structures of practical truth. The articles herewith are submitted as a contribution to this end, and their publication as Bulletin No. 33 of this office is respectfully recommended. Eespectfully, A. C. True, Director. Hon. J. Sterling Morton, Secretary of Agriculture. THE COTTON PLANT. INTRODUCTION. By Chas. W. Dabney, Jr., Ph. D., Assistant Secretary of Agriculture. Cotton is the principal product of eight great States of this Union, and the most valuable "money crop" of the entire country. Climatic conditions practically restrict its cultivation to a group of States consti- tuting less than one-fourth of the total area of our country, and yet the value of the annual crop is exceeded among cultivated products only by corn, which is grown in every state of the Union, and occasionally — four years out of the last ten — by wheat. Cotton furnishes the raw material for one of our most important manufacturing industries and from one-fourth to one-third of our total exports. Considered without reference to any particular country, its economic importance is far beyond numerical expression ; for while the total crop of the world is approximately ascertainable, the effect of cotton upon the commercial and social relations of mankind is too far-reaching for estimation. Of the four great staples that provide man with clothing — cotton, silk, wool, and flax — cotton, by reason of its cheapness and its many excellencies, is rapidly superseding its several rivals. Fifty years ago only about 2,500,000 bales of cotton, or less than the present produc- tion of Texas, were annually converted into clothing; the spindles of the world now use over 13,000,000 bales per annum. Yet less than half the people of the world are supplied with cotton goods made by modern machinery, and Edward Atkinson has estimated that it would require annually a crop of 42,000,000 bales of 500 pounds each to raise the world's standard of consumption to that of the principal nations. Cotton stands preeminent among farm crops in the ease and cheap- ness of its production, as compared with the variety and value of its products. No crop makes so slight a drain upon the fertility of the soil, and for none has modern enterprise found so many uses for its several parts. The cotton plant yields, in fact, a double crop — a most beautiful fiber and a seed yielding both oil and feed, which, although neglected for a long time, is now esteemed worth one-sixth as much as the fiber. In addition to this, the stems can be made to yield a fiber which waits only for a machine to work it, and the roots yield a drug. It is entirely possible, therefore, that cotton may ultimately be grown as much for these parts as for the lint. The history of cotton production in the United States differs from that of almost every other agricultural product in several important 13 14 THE COTTON PLANT. particulars. For nearly three-quarters of a century slave labor was almost exclusively employed iu this branch of agricultural industry, and an immense majority of the colored people of to-day look to it for their chief support. Cotton was also the great pioneer crop in the new south- western States. Not only has the westward movement of the industry been more rapid than that of any other crop, but the center of produc- tion has always been farther in advance of the center of population. As long ago as 1839 Mississippi was producing almost one-fourth of the entire crop of the country. .Recent years have witnessed an enormous development in the regions to the west, which would have carried the center of production across the Mississippi River if the cultivation of cotton, unlike that of wheat and corn and other products, had not taken a new lease of life in the older States along the Atlantic seaboard, where the use of manures has both extended the area and increased the production. Probably no equally great industry was ever more completely paral- yzed or had its future placed in greater jeopardy than cotton growing in the United States during the war of 1861-1865. So great was the decrease in production which followed the effectual closing of the ports that only 1 bale of cotton was grown in 1864-65 for every 15 bales raised in 1861-62. The chief menace to the future of cotton production lay in the efforts that were put forth by other cotton- growing countries at this time to produce those particular varieties which had for so long given the United States the monopoly of the European markets; and nothing could more completely demonstrate the remarkable adaptation of our southern States to the growing of varieties which the experience of generations has proved to be the best for manufacturing purposes than the fact that it took them only thirteen years from the end of the war to regain the primacy of position which they held at its commence- ment. The fact that only a very small fraction of the annual crop fails to reach a market in its raw state explains why we have a more continu- ous and authentic history of the cultivation of cotton in the United States than of any other important product. The ordinary fluctuations to which cotton growing in the United States is subject have been well exemplified in the last ten years, as shown by the following tables: Year. Production (net weight) . Area. Total value of crop. Exports (net weight). Value of exports. 1886-87 3887-88.... 1888-89.... 1889-90.... 1890-91.... 1891-92.... 1892-93.... 1893-94.... 1894-95 1895-96.... Total Pounds. 3, 018, 360, 368 3, 290, 871, 011 3, 309, 564, 330 3, 494, 811, 916 4, 092, 678, 381 4, 273, 734, 267 3, 182, 673, 375 3, 578, 613, 258 4, 586, 594, 540 3, 190, 417, 839 36, 018, 319, 285 Acres. 18, 454, 603 18,641,067 19, 058, 591 20, 171, 896 20, 809, 053 20, 714, 937 18, 067, 924 19, 525, 000 23, 687, 950 20, 184, 368 Dollars. 309, 381, 938 337, 972, 453 354, 454, 340 402, 951. 814 369, 568, 858 326, 513, 298 262, 252, 286 274, 479, 637 287, 120, 818 260, 338, 096 Pounds. 2, 047, 308, 445 2, 150, 775, 786 2, 275, 252, 020 2, 394, 975, 501 2, 759. 049, 246 2, 786, 637, 403 2, 104, 829, 500 2, 547, 624, 248 3, 316, 406, 482 2, 222, 707, 905 Dollars. 206, 222, 057 223. 016 760 237, 775. 270 250, 968, 792 290, 712, 898 258, 461, 241 188, 771, 445 210, 869, 289 204, 900, 990 100, 056, 460 1 19, 931, 539 •Average. 3, 185, 033, 538 24, 605, 566, 536 ; 2, 261, 755, 202 INTRODUCTION. 15 Tear. Average production per acre. Average value per pound. Average value per acre. Per cent of crop ex- ported.1 Per cent of crop retained.1 Pounds. 163. 56 176. 54 173. 65 173. 25 196. 68 206. 31 176. 15 183. 28 193. 63 158. 06 Cents. 10.25 10.27 10.71 11.53 9.03 7.64 8.24 7.67 6.26 8.16 Dollars. 16.76 18.13 18.60 19.98 17.76 15.76 14.51 14.06 12.12 12.90 67.80 65.36 68.75 68.53 67.41 65.20 66. 13 71.19 72.31 69.67 32 20 1887 88 34 64 1888-89 31.25 1889 90 31.47 1890 91 » 32.59 1891-92 34.80 1892 93 33.87 ,i893 94 28.81 1894-95 27.69 1895 96 30. 33 180. 71 8.84 15.98 68.31 31.69 1 Net weight. These tables present some very striking contrasts. The crop of 1894-95 is shown to have been more than half as large again as that of 1886-87, while in respect to value the crop of 1889-90 was just as much in excess of that of 1892-93 or that of 1895-96. While the area cultivated fell nearly 1,900,000 acres below, and rose more than 3,700,000 acres above, the average for the entire period, the wide variation in the average yield per acre played an almost equally important part in determining the total amount of the crop. Yet it is an interesting fact that the fluctuation in yield per acre, taking the country as a whole, is less in the case of cotton than in that of almost any other product of the soil. This is attributable to the greater uniformity of the climatic conditions obtaining in the cotton belt as compared with those of other sections. The rainfall is more constant in the South than in many other sections, the changes of temperature from day to day are con- siderably less than at points in more northern latitudes, and the monthly means of temperatures in the growing season do not vary more than 2° or 3° in the chief cotton-growing districts. In fact, except on the Pacific Coast, there is no portion of the United States where the variability of the temperature is so small, particularly in summer, as in the Southern States and on the Gulf Coast. The increase of sunshine and heat during the months of August and September, con- currently with the decrease of rainfall, which characterize this section so distinctly, are the conditions which favor most the development of the cotton bolls. The same climatic conditions render this section of our country a most favored one for a widely diversified agriculture. The average yield per acre during the ten years covered by the table was 180.71 pounds. Although this may be slightly above the normal yield of the American cotton crop considered as a whole, it can not be doubted that improved methods of cultivation and the increased use of manures are gradually increasing the j>roductiveness of the cotton field, and that the time is not far distant when an average of 200 pounds per acre for the entire cotton belt will no longer excite surprise, while the yield of a bale of 500 pounds will be the standard of the best cotton planters. 16 THE COTTON PLANT. One of the most important facts shown above is that over two-thirds of our entire production of cotton is, and always has been, exported. The time was, indeed, when the proportion shipped abroad was as much as 80 per cent, but the increased production in other countries and the growth of the cotton manufacturing industry in our own have reduced the proportion exported to a little under 70 per cent. One of the serious results of an era of low prices such as that through which the country has been passing is that if it does not, indeed, entirely disable the farmer, it indisposes him to keep up the fertility of his farm. But it is when profits are small that the good results accruing from the adoption of improved methods of cultivation are most apparent. In this connection and at this time, the value to the planter of the work being clone by the various experiment stations and the Department of Agriculture will be thoroughly appreciated. By using, in the most economical form, the fertilizing materials of which the soil is really in need; by disposing the surface of the field and by growing crops in j)roper rotation to prevent the excessive leaching and washing of the soil; by the selection of seed from plants exhibiting the most desirable qualities; by using the most efficient and economical implements and machines for working the soil; by the prompt adoption of the most approved methods of preventing the ravages of injurious insects or the spread of plant diseases, the ordinary farmer inay increase his crop out of all proportion to the additional expenditure bestowed upon it. By using improved gins, adapted to the quality of the cotton grown and the uses to which it is to be put in manufacturing; by packing and handling the cotton in the best manner, and by saving and utilizing the seed or its several products in the most scientific manner, he may get a profit from his crop largely in excess of that ordinarily realized. The demand is not so much for an increased pro- duction, except relatively to the area under cultivation, as for a special- ization of cotton culture and improvements in the methods of ginning and handling the product, so that the most esteemed varieties may be grown in sufficient quantity and at a price that will give the farmer a higher percentage of profit. Wonderful results have been accomplished in the crossing, variation, and general improvement of fruits, flowers, and vegetables, but the great staples, like cotton and wheat, have been comparatively neg- lected in these respects. Our common cultivated species of cotton have already been transformed by peculiarities of soil and climate, by special methods of culture, and by the use of fertilizers from a peren- nial to practically an annual form, and by shortening the period of its growth the limits of successful cotton culture have been greatly ex- tended. But much remains to be done in improving the varieties of, cotton. No plant, however, is more susceptible to such variation and improvement than cotton, and with hundreds of intelligent planters supplementing the work of the experiment stations, results that will be of the greatest benefit to the industry can hardly fail to accrue. HISTORY AND GENERAL STATISTICS OF COTTON. By E. B. Handy, Office of Experiment Stations. ANCIENT HISTORY. The dawn of history shows man using' various fibers for the manu- facture of cloth, and the most ancient traditions of the race prove that the discovery of the art of weaving was made many centuries prior to that time. Wool was the principal material used in Palestine, Syria, Greece, Italy, and Spain; hemp in the northern part of Europe; flax in Egypt; silk in China, and cotton in India; the inhabitants of each of these countries being well skilled in the conversion of these raw fibers into the cloth best suited to their needs. The date at which cotton fiber was first applied to the weaving of cloth by the Hindoos is unknown, but in the digest of ancient laws ascribed to Manu, 800 B. C, cotton is referred to so often and in such a way as to indicate that it must have been known to them for genera- tions, both as a plant and as a textile. The following sentence is quoted from these laws as evidence of the high esteem in which the fiber was held : " The sacrificial thread of the Brahman. must be made of cotton (l-arpasi), so as to be put over the head in three strings; that of the Cshatnya of sana thread, that of the Vaisya of woolen thread." l The following quotation from the same source also bears testimony to the antiquity of weaving and sizing among these people : " Let a weaver who has received 10 palas of cotton thread give it back increased to 11 by the rice water and the like used in weaving; he who does otherwise shall pay a fine of 12 paiias."2 Theft of cotton thread was made punishable by fines of three times the value of the article stolen.3 Herodotus says: "There are trees which grow wild there [India] the fruit of which is a wool exceeding in beauty and goodness that of sheep. The Indians make their clothes of this tree wool."4 His expression " aito $,vkoov 7T€7roi??jn£va,,,:i referring to the clothing of Xerxes's army, is more correctly interpreted "cotton fiber" than the 'Mann, Book II, No. 44. 'Manu, Book VIII, No. 236. 2 Manu, Book VIII, No. 397. "Herod., Ill, 106. •Herod., VII, 65. 1993— No. 33 2 • 17 18 THE COTTON PLANT. fiber of trees (bast fiber). Moreover, the "gvXiva ifxaria^ of the Indians, which is the expression used by Gtesias,1 the contemporary of Herodotus, may be considered as referring to cotton; for Varro, as reported by Servius,2 states: "Ctesias says there are trees in India which bear wool." Theophrastus says: "The trees from which the Indians make cloth have a leaf like that of the black mulberry, but the whole plant resembles the dog rose. They set them in plains arranged in rows so as to look like vines at a distance."3 Strabo, who was most careful and accurate in both his investiga- tions and statements, mentions the trees of India on which wool grew, and says: "The Indians use white raiment and fine white cloths and carpasa."4 The last word is a form of the Sanskrit harpasa, Hebrew Jcarpas, Latin carbasus, and indicates a cloth made of either cotton or flax, Braudes5 and Bitter6 maintaining that it was used by classical writers to denote cotton. Aristobulus, contemporary of Alexander the Great, mentioned the cotton plant under the name of the wool-bearing tree, and stated that the capsule contained seed which were taken out, and that the fiber remaining was combed like wool.7 Nearchus, the admiral of Alexander, who conducted a part of his army down the Indus, around the shore of the Arabian and Persian gulfs to the Tigris, about 327 B. C, says: "There are in India trees bearing, as it were, bunches of wool. The natives made linen garments of it, wearing a shirt which reached to the middle of the leg, a sheet folded about the shoulders, and a turban rolled round the head, and that the linen made by them from this substance was fine, and whiter than any other."8 The Greeks on this expedition made use of an infer- ior kind of raw cotton to pad their clothing and stuff their saddles.9 Pliny1" writes of the cotton plant in India with leaves similar to the mulberry and resembling the dog rose, which was sowed in the field, and from which the inhabitants made linen clothes. Here Pliny uses the word Tineas, but the context shows that he referred to cotton. Quin- tus Curtius says of the Indians : " They covered their bodies from head to foot with carbasus; they bind shoes about their feet, linen cloths about their heads;" and, speaking of the dress of the king, he says: "The carbasa which he wore were spotted with purple and gold."11 fragment of Ctesias (ed. Muller, p. 84). 2 Coram, in Virgilii JEn., I, 649. 3Tkeoph. Hist. Plant., IV, 4 (p. 132, ed. Schneider). 4Strabo, XV, 719. 5Ueber die antiken Namen nnd die geog. Verbreit. der Bannrwolle im Alterthum, p. 107. «Geog. Vol. IV, 1, p. 436. 7Strabo, XV, 1 (Vol. VI, p. 43, ed. Siebenkees). 8 Arrian. Ind., ch. 16. n Strabo, XV, 693. 10Plin. Hist. Nat., XII, 13. » Luc., VIII, 9. HISTORY AND GENERAL STATISTICS OF COTTON. 19 Lucau,1 the poet, who lived in the first century of the Christian era, describes the inhabitants of India as those : Who drink sparkling juices from tender cane, With dyes of crocus stain their hair, and fix With colored gems the flowing carbasus. The Periplus 2 states that two kinds of cotton goods — a fine and a coarse grade — were sent from Ariaca and Barygaza (the modern Baroack on the Nerbudda) to Malao, Mundi, Mosyllou, Tabae, to Opone on the east coast of Africa, and to the islands lying off that coast. And the following places are mentioned in the same book3 as manufacturing centers for the same cloth : Palasimnnda, Masalia (the modern Masuli- patam, on the east coast of India), and the country around the mouth of the Ganges. These references demonstrate that cotton was not only known and used by the Hindoos in early times, but that they manufactured cloth from the fiber in sufficient quantities to supply their own needs and to export, or rather sell to traders, who carried it to other places. The cultivation of the plant or the manufacture of the cloth from the fiber did not receive as much attention in any country as in India. In fact, from 1500 B. 0. until an equal number of years after the beginning of the Christian era, India was the center of the cotton industry. The cotton cloth which the Indians produced from a short fiber with primitive distaffs and rude looms was not equaled until the last half century. Some of their muslins possessed wonderful delicacy of texture. Two Arabian travelers of the middle ages, writing of India, say: "In this country they make garments of such extraordinary perfection that nowhere else are the like to be seen. These garments are for the most part round, and woven to that degree of fineness that they may be drawn through a ring of moderate size."4 Marco Polo,5 whose work was first circulated about 1298 A. D., men- tions the coast of Coromandel as producing "the finest and most beau- tiful cottons that are to be found in any part of the world." Odoardo Barbosa,6 one of the Portuguese who visited India imme- diately after the discovery of the passage around Good Hope, speaks of "the great quantities of cotton cloths admirably painted, also some white and some striped, held in highest estimation," which were made in Bengal; and Csesar Frederick7 mentions cotton cloth so valuable "that a small bale of it will cost 1,000 or 2,000 duckets." 1 Luc, III, 239. 2 Periplus maris Erythrsei, n 8, 9, 10, 12, 13, 14, 31. 3Loc. cit..U 51,61,63. 4Anciennes Relations des Indes et de la Chine, etc., p. 21. fi Travels of Marco Polo, Book III, ch. 21, 28. 6Ramusio's Raccolta delle Navigationi et Viaggi, Tome I, p. 315. 7Hakluyt's Voyages, Vol. II, p. 366, ed. 1809 20 THE COTTON PLANT. Tavernier1 says "some calicuts are made so fine you can hardly feel them in your hand, and the thread when spun is scarce discernible;" also, that "the rich have turbans of so fine a cloth that thirty ells of it ■put into one turban make it weigh less than four ounces." "When the muslin is laid on the grass to bleach and the dew has fallen upon it, it is no longer discernible," says Ward;2 and an English review' of the trade of the latter part of the seventeenth century designated the same fabrics as but "the shadow of a commodity." Surely the poetic writers of the Orient w ere justified in calling them "webs of woven wind." Cotton was introduced into China and Japan from India, but com- mon use of the fiber in these oriental countries, where at the present time it is even more widely used than in the West, came as slowly as the adoption of it as clothing among Europeans, and met with equally active opposition. Although China carried on an exchange of prod- ucts with India in very early times, both by way of caravans and by junks coasting along the shores, it was as late as the thirteenth century of the present era before her inhabitants began the cultivation of cot- ton as anything but a garden plant. Marco Polo,4 although several years a resident of China, with every facility for observation, gives no account of cotton culture except in the province of Fo-Kien, but speaks of silk as the usual dress of the people. It appears, however, from Chinese history that the plant had been known for many centuries, but that its manufactured product was a rarity. The introduction of cotton into China dates practically from the con- quest of China by the Tartars, and it was not until 1300 A. D. or thereabouts that it was cultivated for general use. The cotton plant also thrived in ancient times in the island of Tylos, situated near the Arabian coast in the Gulf of Persia. Theophrastus5 mentions wool-bearing trees which grew abundantly in this island, and which had leaves like those of the vine, but smaller. They bore a capsule about the size of a quince, which, when ripe, burst, disclosing the seed surrounded with a wool, which was woven by the people into cloth of different qualities. Pliny fi makes similar statements regarding the wool-bearing trees of this island. The climate and soil of the island of Tylos were so favorable to the plant that cloth made of the cotton of Tylos was preferred to that of India.7 Although there is no mention of cloth being made in Tylos, it probably was made there, and also on the neighboring mainland. ^avernier's Travels (Harris's Collection of Voyages, Vol. I, p. 811). 2View of the History, Literature, and Mythology of the Hindoos, Vol. Ill, p. 127, 3d ed. 3The Naked Truth, p. 11. "Travels, Book II, ch. 74. 5Theoph. Hist. Plant., IV, 7, 7. 6Plin.Hist.Nat.,XII,21. 7 Plin. loc. cit., XII, 22. HISTORY AND GENERAL STATISTICS OF COTTON. 21 Because Pliny describes the Arabian tree from which the fiber was secured to make a cloth similar to linen as having- leaves like the palm,1 some have doubted the presence of the true cotton plant in Arabia. But Theophrastus2 says Arabia is a producer of the same plant, which he had just described as a product of Tylos; and Pliny, in another connection,3 describing the different kinds of cotton plants, intimates that besides the variety with the palm-like leaves, another, probably the same plant described by Theophrastus, also grew there. The Peri plus4 mentions Omana, in the southern part of Arabia, as a place from which cotton cloth was shipped, not only to other portions of Arabia, but even to India. The cotton plant was known on the eastern coast of Africa as well as in the southern parts of Asia. Although much has been written to prove that the ancient Egyptians used only flax for weaving and knew nothing of the cotton plant, it seems most probable that this opinion is based mainly upon a too narrow interpretation of the terms used by classical authors and the idea that both flax and cotton were never used by the same people. The fact is that both flax and cotton were used, alone and mixed, by both the Egyptians and Indians. Egypt was one of the most ancient as well as most populous empires of antiquity, and her inhabitants were early compelled to turn their attention to other than agricultural occupations; so that the various industries were well known from the date of the earliest traditions, and among them none were more developed than weaving. While flax was probably the most common article used by Egyptian weavers in the manufacture of cloth, and linen was in fact the material of which the clothing of the people and the wrappings of their dead were usually made, it appears quite arbitrary to state that the Egyptians knew noth- ing of cotton, and consequently made no use of it in ancient times, for the probability is that it was through the commercial and industrial activity of this people that cotton was brought to the shores of the Mediterranean Sea.5 The Egyptians were adventurous sailors, and had early established a trade from ports on the Red Sea with those on the western coast of India. Doubtless they brought home many of the fine fabrics made by the skillful weavers of India from the fiber of the cotton plant. This cotton plant also grew in Egypt, for Pliny says: "In upper Egypt, toward Arabia, there grows a shrub which some call 'gossypion' and others 'xylon,' from which the stuffs are made which Ave call 'xylina.' It is small and bears a fruit resembling the filbert, within which is a downy wool which is spun into thread. ^Nothing is more to be desired than this goods for whiteness and softness. Garments are made from it which are very acceptable to the priests of Egypt."0 And Pollux, who 1 Plin. Hist. Nat., XII, 22. ^ Periplus, vS 36. JTheoph. Hist. Plant., IV, 7, 8. HBrandes, loc. cit., p. 111. ■spiiu. Hist. Nat., XIX, 3. 6Plin. Joe. cit., XIX, 3. 22 THE COTTON PLANT. lived about 150 A. D., reports that " among the Indians, and now also among the Egyptians, a sort of wool is obtained from a tree. The cloth made from this wool may be compared to linen, except that it is thicker The tree produces a fruit most nearly resembling a walnut, but three- cleft. After the outer covering has divided and become dry the sub- stance resembling wool is extracted and is used in the manufacturing of cloth for weft, the warp being of linen."1 These two passages taken together are worthy of consideration, and no mere dispute as to the meaning of the terms used by other writers can throw discredit on such distinct statements. The description Pol- lux gives of the cotton tree is remarkably correct — more so than any given before his time, unless it be that of Aristobulus and Nearchus, the companions of Alexander the Great. The description of the pericarp as three-cleft and the comparison of the boll to a walnut are in striking agreement with the facts. Besides, he goes into details as to the use of the thread for the weft only, while use of linen thread for warp is stated. Although Herodotus states, in referring to the Egyptian priests, that they wore linen clothes,2 Pliny, as above quoted, says that cotton clothes were very acceptable to them, and Philostratus 3 confirms his statement. Just here it may be well to remark that the word translated "linen" did not always refer to the fiber of which a material was made, but often to the general appearance of the cloth ;4 therefore cloth made of either flax or cotton alone or mixed is called linen ; so that the direct statements of these authors are not necessarily contradictory. Moreover, the use of cotton, a vegetable fiber, in the garment of the priests would be considered equally as pure as that of linen, because it would not be in conflict with the religious rules which proscribed wool, an animal product. The fact that all mummy cloths, so far as yet examined, have been found to consist of flax has been used as a basis for an argument to prove that cotton was not used by the Egyptians; but it seems far from conclusive, as for religious reasons flax alone may have been used for that purpose, while cotton, wool, and silk may have been regularly used by the living for clothing or ornament. Elsewhere it is even stated that the form and manner of wearing these clothes was like a shawl or mantle,5 and that the Egyptian cotton cloth was thicker than that made of linen and was embroidered.6 In reference to the lands on the west coast of Asia, there are so many traces of the use of cotton, especially by the Semitic tribes of antiquity, 1 Polluc. Onomast., VII, 75. " Herod., II, 37. 3Philostr. Vit. Apollon., II, 9. 4Muller, Handbucli o that the crop was a failure that year. Our first supposition was that the cause was one of price, hut on examining the quotations in Took's work on "high and low prices" we do not see any marked decline in the values of other descriptions of cotton, and the American staple is not given in his list until 1793. 1791. — One hundred and eighty-nine thousand live hundred (189,500) pounds imported into Liverpool, the price averaging here 26 cents. 1792. — One hundred and thirty-eight thousand three hundred and twenty-eight (138,328) pounds imported into Liverpool. Great difficulty was experienced in separating the seed from the lint of upland cotton. The work was done by hand, the task being- 4 pounds of lint cotton per week from each head of a family, in addition to the usual field work. This would amount to one bale in two years. A French planter of Louisiana (Dubreuil) is said to have invented a machine for separating lint and seed as early as 1742. The demand for such a machine not being very great at that date, no record as to its character has been preserved. The roller gin, in very much the same form as IsTearchus, the admiral of Alexander the Great, found it in India, was still in use. In 1790 Dr. Joseph Eve, originally from the Bahamas, but then a resident of Augusta, Ga., made great improvements on this ancient machine, and adapted it to be run by horse or water power. A correspondent of the American Museum, writing from Charleston, S. C, in July of that year, states " that a gentleman well acquainted with the cotton manufacture had already completed and in operation, on the high hills of Santee, near Statesburg, ginning, carding, and other machines driven by water, and also spinning machines witli eighty-five spindles each, with every article necessary for manufacturing cotton." A machine dating anterior to this year, and having a strong resem- blance to the above, possessing in fact all the essentials of a modern cotton gin, was exhibited at the Atlanta Exposition in 1882. It came from the neighborhood of Statesburg, but its history could not be ascertained. In 1793 Eli Whitney petitioned for a patent for the invention of the saw cotton gin. His claims were disputed, and he defended tbem in the State and Federal courts for nearly a generation, obtaining at last a verdict in his favor. Meanwhile the saw gin had become an estab- lished fact, and the planter at last had a machine which enabled him to produce cotton at a cost that would leave him a good profit. The first saw gin to be run by water power was erected in 1795 by James Kiucaid near Monticello, in Fairfield County, S. C. Others were put up near Columbia by Wade Hampton, sr., in 1797, and in the year follow- ing he gathered and ginned from 600 acres 600 bales of cotton. The cotton exportation from the United States increased from 487,600 pounds in 1793 to 1,600,000 pounds in 1794, the year in which Whitney's gin was patented. In 1796, a year after he had improved his machine, the production had risen to 10,000,000 pounds. In fact, the increased production was so great that the planters began to fear they would HISTORY AND GENERAL STATISTICS OF COTTON. OV overstock the market, and one of tbern, upon looking at liis newly gath- ered crop, exclaimed: "Well, I liave done with cultivation of cotton; there's enough in that ginhouse to make stockings for all the people in America." Yet the production of cotton did not advance with that rapidity to which Ave are now accustomed. The cotton industry being of secondary importance prior to 1790, information and statistics relative to the amount produced are not avail- able, but the following table ' gives the production at different dates from that time to 1895 : Cotton crops of the United States at stated periods. [Expressed in bales of 400 pounds net.] 1790-91 1791-92 1796-97 1800-01 1801-02 1806-07 1811-12 1816-17 1821-22 1826-27 1834-35 1840-41 1850-51 1855-56 Season. 1859-60 1865-66 1869-70 1873-74 1878-79 1880-81 1884-85 1887-88 1890-91 1891-92 1893-94 1894-95 1895-96. Bales. 602, 639 501, 921 434, 806 629,131 670, 368 596, 613 562. 090 227,178 231, 696 684, 336 946, 533 406, 486 976, 045 Within one hundred years, from 1790 to 1S90, the production of cot- ton in the United States increased from 5,000 bales to over 10,000,000 bales. The crop of 1891-95 exceeded that of 1878-79 by nearly 6,000,000 bales, or over 100 per cent increase in sixteen years. The following tables show the cotton acreage for the years 1878 to 1896, inclusive, and the estimated production and exportation of Sea Island cotton for the years 18S0 to 1894, inclusive: Acreage in cotton in the United States : 1878 to 1896. Season. North Carolina. 1878-79 ... 1879-80 1880-81 1881-82 1882-83 1 1, 050, 543 1883-84 1 1, 050, 543 1884-85 ! 1, 061, 048 1885-86 - 1. 071, 058 icres. 590, 500 625. 900 933, 000 1, 061, 155 1886-87. 1887-88. 1888-89. 1889-90. 1890-91 . 1891-92. 1892-93. 1893-94. 1894-95. 1895-96. 1896-97. 1, 071, 658 1, 006, 301 1,071,633 1.147, 136 1, 036, 026 1,014,270 833, 695 1, 180.000 1, 296, 522 1.050, 183 1,228,714 South Carolina. Acres. 944, 650 944, 600 1, 441, 600 1, 619, 639 1, 618, 989 1,618,989 1, 716, 128 1, 733, 289 1,655,291 1, 622, 185 1, 640, 518 1, 987, 469 1, 682, 522 1, 663, 997 1, 482, 222 1, 885, 000 2, 160, 391 1,814,728 2, 014. 348 Georgia. Florida. ; Alabama. Missis- sippi. Acres. 1, 483, 500 1, 592, 000 2, 786, 300 2, 994, 005 2, 872, 748 2, 872, 748 2, 958, 930 3, 047, 698 2, 956, 267 2, 941, 486 2, 970, 901 3, 345, 104 3, 086, 682 2, 991. 027 2, 841, 580 3, 050. 000 3, 610. 968 3, 069, 323 3, 468, 335 Acres. 166. 650 161.600 251. 600 263, 032 257. 799 257, 799 268. Ill 273, 473 270, 738 262, 616 259, 990 227, 370 280, 147 271, 467 214, 017 165, 000 201, 621 191,540 264, 325 Acres. 1, 837, 550 1, 892. 700 2, 460, 600 2, 639, 988 2,610,420 2, 610, 420 2, 740, 941 2, 795, 760 2, 823, 718 2, 809, 599 2,851,743 2, 761, 165 3. 096, 232 3, 000, 282 2, 607. 616 2,316,000 2, 664, 861 2.371,720 2, 656, 333 Acres. 2, 055, 050 2, 055, 000 2, 275, 000 2, 351, 228 2, 278, 521 2, 278, 521 2, 392, 447 2, 535, 994 2, 548, 674 2, 548, 674 2, 592, 001 2, 883, 278 2, 965, 614 2, 903. 327 2, 571, 684 2. 845, 400 2, 826, 272 2. 487. 119 2, 835, 316 1 Except where otherwise stated, the statistics used in this article are taken mainly from Rpt. IT. S. Senate Com. on Agr. and Forestry, Feb. 23, 1895; Snepperson's Cotton Facts, 1895, and reports of the Bureau of Statistics, U. S. Treasury. 40 THE COTTON PLANT. Acreage in cotton in the United States: 1878 to 1S9G— Continued . 1878 79 1879-80 1880-81 1881-82 1882-83 1883-84 1884-85 1885-86 1886-87 1887-88 1888-89 1889-90 1890-91 1891-92. 1892-93 1893-94 1894-95 1895-96. 1896-97 Louisiana Acres. 1, 348, 950 1,322,000 888, 000 944, 174 931,900 931,900 922, 581 1,005,613 1, 035, 781 1,066,854 1,088,191 1,270,154 1,107,138 1, 094, 949 897, 469 946, 000 1,313,296 1, 142, 568 1,245,399 Texas. Acres. 1, 808, 400 1,935.000 2, 395, 100 2. 676, 298 3. 034, 922 3. 034, 922 3, 186, 668 3, 505. 335 3,771,740 3, 960, 327 4,158,343 3, 934, 525 4, 956, 145 5, 198, 643 4,431,729 4, 153. 760 6, 854, 621 5, £26, 428 6, 758, 656 Arkansas. Acres. 1,165,850 1, 177, 500 1,080,200 1,181,692 1,188.545 1, 188, 545 1, 259, 858 1,348 048 1,354,788 1, 388, 658 1,416,431 1,700,578 1,494,333 1, 492, 809 1, 247, 239 1,867,250 1,483,319 1, 186, 655 1,542,652 Tennessee. A cres. 740. 700 702, 900 816, 200 840, 990 807, 602 807, 602 815, 678 864, 618 847, 326 855, 799 881,473 747,471 995, 181 964, 341 808, 235 805, 920 879. 954 712.763 912,337 All other. A cres. 125,000 120, 300 147, 700 138, 529 139, 508 126, 004 117, 222 119,379 118,622 118, 568 121, 367 167, 646 109, 033 119, 825 132, 438 310, 670 396, 125 331, 335 518,919 Total. Acres. 12, 266, 800 12, 595, 500 15, 475, 300 16, 710, 730 16, 791, 557 16,777,993 17,439,612 18, 300, 865 18,454,603 18, 641, 007 19, 058, 591 20, 171, 896 20, 809, 053 20, 714, 937 18, 067, 924 19, 525, 000 23, 687, 950 20, 184, 368 23, 445, 334 Estimated Sea Island cotton crop of the United States and exjwrts of same since 1880. Year. 1880-81 1881-82 1882-83 1883-84 1884-85 1885-86 1886-87 1887-88 1888-89 1889-90 1890-91 1891-92 1892-93 1893-94 1894-95 Florida. Georgia. Carolina. Texas- Total- Exports. Bat 16, 20, 16, 16, 23, 23, 29, 22, 22, 23, 22, 17, 9, 19, 15, Hales. 3, 179 6,049 3,126 1,399 4,327 5,780 6,411 8,304 12. 000 13, 629 29, 613 30, 576 28, 324 39, 367 53, 703 South Bales. 14. 868 10,796 10,591 7,329 12, 588 8,497 8,735 8,561 9,618 9, 256 16. 306 11,499 7,212 2,578 5,894 Bales. 35, 021 37, 862 36, 709 25. 490 40, 452 37, 778 45, 137 39, 479 44, 089 46, 803 68, 133 59, 134 45.418 61, 052 74, 628 Bales. 24, 395 24, 756 23, 457 13, 579 21, 565 16, 428 26, 651 20, 613 23, 326 28, 242 39, 123 27, 431 22, 540 38.021 40, 744 The first cotton mill erected in the United States was built at Bev- erly, Mass., in 1 787-88. This was soon followed by others in various towns along the east border of the country, especially Pawtucket and Providence, P. L, Boston, Mass., New Haven and Norwich, Conn., New York City, Paterson, N. J., Philadelphia, Pa., and Statesburg, S. O. In them carding and spinning were done by machinery, but the weav- ing was on hand looms until 1815, at which date a power-loom mill was started at Waltham, Mass. The use of hand looms and spinning wheels for cotton manufacture was common in all parts of the country before the Eevolution, especially in the Southern colonies, and these continued to be used by the women in their houses many years after the erection of cotton factories. * In 1831 there were in the United States 801 cotton mills, with 33,433 looms and 1,246,703 spindles, employing 62,208 persons and consuming 77,457,316 pounds of raw cotton, with $40,612,984 capital. In 1860 there were 1,091 mills, with 126,313 looms and 5,235,727 spindles, employing 122,028 workmen, and having $98,585,269 capital. The raw material used was 422,704,975 pounds, valued at $57,285,534, and the value of the finished product was $115,681,774. HISTORY AND GENERAL STATISTICS OF COTTON. 41 During the great "cotton famine" caused by the civil war the pro- duction of cotton in the United States practically ceased, and it was not until 1867-68 that the cotton industry regained the position it had held in 1860. From that period until 1880 there was irregular but decided growth in the industry. The supply and consumption of cotton in the United States from 1791 to 1895 is shown in the following table, compiled by Watkins:1 Production and consumption of cotton in the United States from 1791 to 1S05, inclusive. 1790-1791. 1791-1792- 1792-1793. 1793-1794. 1794-1795. 1795-1796. 1796-1797. 1797-1798. 1798-1799. 1799-1800. 1800-1801. 1801-1802. 1802-1803. 1803-1804. 1804-1805. 1805-1806. 1806-1807. 1807-1808. 1808-1809. 1809-1810. 1810-1811. 1811-1812 1812-1813. 1813-1814. 1814-1815. 1815-1816. 1816-1817. 1817-1818. 1818-1819. 1819-1820. 1820-1821. 1821-1822. 1822-1823. 1823-1824. 1824-1825. 1825-1826 1826-1827. 1827-1828. 1828-1829. 1829-1830. 1830-1831. 1831-1832. 1832-1833 1833-1834. 1834-1835. 1835-1836. 1836-1837. 1837-1838. 1833-1839. 1839-1840. 1840-1841. 1841-1842. 1842-1843. 1843-1844. 1844-1845. Crop. Bales. 8,889 13, 333 22, 222 35, 556 35, 556 44, 444 48, 889 66, 667 88, 889 155,556 210, 526 241, 228 252, 101 240, 741 281,128 347, 826 285, 714 271,739 366, 071 340, 000 269, 360 304, 878 304, 878 284. 553 363.636 457, 565 460, 993 448, 029 596, 429 606. 061 647, 482 742, 049 620, 805 762,411 891,608 1,121,667 957, 281 720, 593 870, 415 976. 845 1,038,847 987. 477 1, 070. 438 1,205,394 1,254,328 1,360,725 1,423,930 1,801,497 1, 360, 532 2, 177, 835 1,634,954 1,683,574 2,378,875 2, 030, 409 2, 394, 503 1845-1846 i 2,100,537 1840-1847 1 1,778.651 1847-1848 ( 2,439,786 1848-1849 [ 2,866,938 aXo data. Consump- tion. Bales. (a) («) (a) (a) («) (a) («) («) (a) 35, 556 39, 474 (a) (a) (a) 44,177 («) («) (a) (a) 64, 000 57, 239 (a) («) (a) 90, 000 (a) (a) (a) (a) (a) (a) (a) (a) (a) (a) la) 149, 516 120, 593 118,853 126, 512 182, 142 173, 800 194, 412 196,413 216. 888 236, 733 222, 540 246, 063 276, 018 295, 193 297, 288 267, 850 325, 129 346, 750 389, 000 422, 600 428, 000 616,044 642, 4S5 Stock Exports. (close of year). Bales. Bales. 889 (a) 635 (a) 2 222 (a) 7,407 (a) 27, 822 (a) 27, 141 (a) 16, 837 (a) 41, 600 (a) 42, 366 (a) 79, 066 («) 91,716 (a) 120, 619 W 158, 454 (a) 129, 756 (a) 154, 101 (a) 155, 032 (a) 228, 362 (a) 38, 516 (a) 227, 635 (a) 373, 046 (a) 208. 950 (a) 117,428 (a) 77, 683 («) 72, 069 (a) 301,814 (a) 302, 388 (a) 303, 721 (a) 331,438 (n) 314, 275 (") 484, 319 («) 449, 257 (a-) 511,219 (a) 582, 964 (a) 504, 857 (a) 616, 958 («) 655, 562 (a) 854, 000 (a) 600, 000 40, 000 740, 000 30, 000 839, 000 35, 000 773, 000 119,000 892, 000 41, 600 867, 000 48, 200 1, 028, 000 29, 600 1, 023, 000 41, 600 1, 116, 000 43, 300 1,169,000 75, 800 1, 575, 000 40, 300 1, 074, 000 52, 250 1, 876, 000 58, 442 1, 313, 500 72, 479 1,465,500 31. 807 2,010,000 94, 480 1,629,500 159. 772 2, 0S3, 700 98, 420 1,666,700 107, 122 1,241,200 214, 837 1, 858, 000 171,468 2, 228, 000 154,753 Net weight of bales. Pounds. 225 225 225 225 225 225 225 228 228 238 270 249 230 280 276 224 250 297 246 246 246 275 271 282 279 280 264 278 283 298 282 286 312 331 335 341 339 341 360 350 363 367 373 379 379 384 383 394 397 409 412 415 411 436 411 437 1 Production and Price of Cotton for One Hundred Years, IT. S. Dept. Agr., Div. of Stat. Misc. Bnl. 9. 42 THE COTTON PLANT. Production and consumption of cotton in the Ignited States, etc. — Continued. Tear. 1849-50 1850-51 1851-52 1852-53 1853-54 1854-55 1855-56 1856-57 1857-58 1858-59 1859-60 1860-61 1861-62 1862-63 1863-64 1864-65 1865-66 1866-67 1867-68 3868-69 1869-70 1870-71 1871-72 1872-73 1873-74 1874-75 1875-76 1876-77 1877-78 1878-79 1879-80 1880-81 1881-82 1882-83 1883-84 1884-85 1885-86 1886-87 1887-88 1888-89 1889-90 1890-91 1891-92 1892-93 1893-94 1894-95 Crop. Bales. 2, 333, 718 2, 454, 442 3,120,310 3, 416, 214 3, 074, 979 2, 982, 634 3, 655, 557 3, 093, 737 3, 257, 339 4,018,914 4,861,292 3, 849, 469 0,4, 500, 000 al, 600, 000 b 450, 000 6300,000 2, 269. 316 2, 097, 254 2, 519, 554 2, 366, 467 3, 122, 551 4, 352, 317 2, 974, 351 3, 930, 508 4, 170, 388 3,832,991 4, 632, 313 4, 474. 069 4, 773,' 865 5, 074, 155 5,761,252 6, 605, 750 5, 456, 048 6, 949, 756 5, 713, 200 5, 706, 165 6, 575, 691 6, 505, 087 7, 046, 833 6 938,290 7, 311, 322 8, 652, 597 9, 035, 379 6, 700, 365 7, 549, 817 c9, 476, 435 Consump- tion. Bali's. 613, 498 485, 614 689, 603 803, 725 737, 236 706,417 777, 739 819, 936 595, 562 927, 651 978, 043 843, 740 o370, 000 a288, 000 «220, 000 0345, 000 666, 100 770, 030 906, 636 926, 374 865, 160 1, 110, 196 1, 237, 330 1,201,127 1, 305, 943 1,193,005 1,351,870 1,428,013 1,489,022 1,558,329 1,789,978 1,938,937 1, 964, 535 2, 073, 096 1, 876, 683 1, 753, 125 2, 162, 544 2,111,532 2, 257, 247 2, 314, 091 2,390,959 2, 632, 023 2, 876, 846 2, 431, 134 2, 319, 688 2,704,153 Stock Exports. (close of year). Bales. Bales. 1, 590, 200 167,930 1,988,710 128, 304 2, 443, 046 91, 170 2, 52K, 400 135, 643 2, 319, 148 135, 603 2, 244, 209 143, 330 2, 954, 606 04, 171 2, 252, 657 49, 258 2, 590, 455 102, 926 3,021,403 149, 237 3, 774, 173 227, 708 3, 127, 568 83, 127 ; 644, 936 (b) 10, 898 (b) 27, 053 (b) 24, 787 (b) 1,554,664 283, 692 1, 557, 054 80, 296 1, 655, 816 37, 398 1,465,880 11,160 2, 206, 480 65, 325 3, 109, 009 144, 290 1,957,314 59, 287 2, 679, 986 104,782 2, 840, 981 124, 795 . 2, 684, 708 74,411 3, 234, 244 130,041 3, 030, 835 130, 493 3, 360, 254 45, 784 3,481,004 65, 948 9, 885, 003 141,418 4, 589, 346 218, 043 3, 582, 622 124, 232 ! 4, 766, 597 237, 117 3,916,581 116, 190 3, 947, 972 132, 421 4, 336, 203 178, 026 4, 445, 302 86, 269 4, 627, 502 180, 062 4, 742, 347 65, 624 4, 906, 027 75, 195 5, 847, 191 215, 692 5, 933, 437 421,104 4,445,338 237, 411 5, 287, 887 180,912 6, 614, 619 405, 519 a Estimated. b Xo data. c Estimate of Department of Agriculture. The exports of cotton from the United States to Great Britain from 1780 to 1790 averaged only ^ of the total cotton imports of that coun- try, but sixty years later, from 1816 to 1S50, the United States supplied four-fifths of Great Britain's demand for cotton. From 1780 to 1790 the average amount of cotton imported by Great Britain from the United States was 100 bales (of 100 pounds each); from 1816 to 1820, 166,310 bales; from 1846 to 1850, 1,297,230 bales ; and from 1876 to 1880, 2,589,070 bales. Of the supply of cotton available for the United States and Europe in 1850-51 the United States furnished 2,536,000 bales, or 81 per cent, and although the gross production has so greatly increased since that date that in 1891-92 the United States produced 10,661,000 bales, the proportion of the cotton available for Europe and the United States furnished by the United States has never been greater than 82 per cent (in 1892). Price in cents perlb. middling in New York. "K ■K S) 5 _1 X < 26. 28. 16. 14. 21. 17. 8. 12.5 10.04 6. 13.5 8.92 8.5 1.3. 11.21 10.25 13.87 12.34 10.75 11.82 11. 70. 187. 101.50 26.5 42. 31.59 14.87 21.13 16.95 10.5 13. 11.35 9.5 11.5 10.71 10.25 11.62 11.53 7.93 10.62 9.03 6.69 8.75 7.64 6.18 10. 8.24 6.87 6.56 7.67 5.56 7.37 6.26 , Expt. Sta. Bui, 33. Production and Consumption of Cotton in bales of 400 lb. weight by the Countries contributing to the world's supply and demand for a series of years from 1790 to 1895. By Harry Hammond. inch. Scale: i.ooo.c ales of 400 lb CONSUMPTION M B r Ch. nr — ' UNITED STATES, NORTH UNITED STATES, SOUTH GREAT BRITAIN CONTINENT OF EUROPE INDIA VARIOUS G C HDoc 267 54 2 HISTORY AND GENERAL STATISTICS OF COTTON. 43 The following table shows the exports of cotton from the United States for the years noted : Exports of cotton from the United States to Great Britain and the Continent and JFexico from 1S79 to 1S94, inclusive. Tear. To Great Britain. To the Continent and Mexico. Total. Tear. To the To Groat , Continent „, + , Britain. and iotal- ! Mexico. 1879-80 1880-81 1881-82 1882-83 1883-84 1884-85 1885-86 1886-87 Bales. 2, 053, 000 2, 554, 000 2, 832, 000 2, 295, 000 2, 886, 000 2, 485, 000 2, 425, 000 2, 565, 000 2, 704, 000 Bales. 1, 413, 000 1, 310, 000 1, 733, 000 1, 256, 000 1,838,000 1,432,000 1, 495, 000 1, 771, 000 1, 741, 000 Bales. 3, 466, 000 3, 864, 000 4, 565, 000 3, 551, 000 4, 724, 000 3, 917, 000 3, 920, 000 4, 336, 000 4, 445, 000 1887-88 1888-89 1889-90 1890-91 1891-92 1892 93 1893-94 1894-95 1 Bales. Bales. Bales. 2,814,000 ! 1,813,000 4,627,000 2, 810, 000 1, 926, 000 4, 736, 000 2,854,000 : 2,052,000 4,906,01)0 3,345,000 1 2,446,000 5,791,000 3, 317, 000 ' 2, 541, 000 5, 858, 000 2,301,000 1 2,089,000 , 4,390,000 2, 861, 000 2, 371. 000 5, 252, 000 3,449,000 3,279,000 6,726,000 COTTON IN INDIA. India ranks, and perhaps always will rank, next to the United States as a cotton-producing country. With an area of 1,307,000 square miles, lying south of the thirty-fifth degree of north latitude and wholly within the cotton belt, India is twice the size of that part of the United States known as the Cotton States, and possesses a good cotton soil, although hampered by an uncertain and discouraging climate. Bounded on the east, north, and west by mountains, with mountain chains traversing the central territory, and subject to two periodical wet seasons, portions of her territory are rendered unfit for cotton growing, either by excessive rainfall, which in some sections amounts to 500 inches per annum, or by the lack of moisture in others, where the annual rainfall is scarcely an inch. Although cotton had been cul- tivated there for fully 4,000 years, the increase in production was but slight until stimulated by the diminished supply from the United States between 1801 and 1865. During the cotton famine of this period the cultivation was pushed to its utmost extent, but when the United States regained its supremacy in cotton culture the production of cotton in India was not pressed with so much vigor. At present the attention of the ryots has been turned to the production of the more profitable indigo and linseed, and it is probable that the production of cotton will further decrease. The average yield in India varies in the different provinces from 40 to 100 pounds of clean cotton per acre, dependent on the seasons. British India, or Hindostan, the part of India where cotton is raised, embraces four principal cotton regions: The valley of the Ganges, the Deccau, western India, and southern India. The Ganges Valley is again divisible into two parts, the Lower Ben- gal district, and that of the northwest provinces, including Doab and Bundelcund, lying on both sides of the Ganges and Jumna rivers. In Lower Bengal the cultivation of cotton is not of very great impor- tance. In the plains of Bengal, which are so fertile in other produce, the production of cotton is very inconsiderable, and none is exported. 44 THE COTTON PLANT. The cotton raised here in former times, though short in staple, was the finest known in the world, and formed the material out of which the very delicate and extremely beautiful Dacca muslin was manufactured. This interesting and indefinite variety of Gossypium kerbaceum is known as Dacca cotton, and what little is raised is used at home in the looms of a few weavers at Bazitpore and Polsia and seldom finds its way to Calcutta. The border lands of the Ganges are too low and marshy and the rain- fall too great for the successful cultivation of cotton, but the hills back from the river are suitable for this purpose, as they are better drained. The Doab and Bundelcund districts produce almost the entire crop of the northwest provinces, and furnish about 70,000,000 pounds of cotton for exportation, which is a good " India cotton." The climatic character of these districts is "first a flood and then a drought," with an inclina- tion to an insufficiency of rain, in great contrast to that of Lower Bengal. The Deccan, or Central India, is the great cotton section of India. It occupies the triangular area lying south of the Vindhyan Moun- tains, in latitude 23° north, and extends to the valley of the Kistna, at 10° north, with the Eastern and Western Ghauts on either side. It is an elevated table-land of undulating surface, having soil of great excel- lence and richness and of a consistency to retain moisture for a long ime. Nearly all the cotton for export is raised within this region and finds its market at Bombay. The Deccan may be divided into the Nagpore, Hyderabad, Berar, and Dharwar districts. The soil in the valleys of Nagpore is a rich black loam which becomes very sticky and muddy during the rainy season and hard and cracked during the dry season, in this respect very much resembling some of the Alabama soils. In the hilly portion there is a red clay soil. The cotton grown within this district is very fine and soft, indicative of a moist and equable climate, especially that produced in the valleys of the Wurda and its tributaries. It is known commercially as "Hingaughat cotton," from the chief town of that section, and is considered as pos- sessing the highest qualities of any India cotton. Hyderabad is a plateau with a surface more or less hilly and a gen- eral elevation of 2,000 feet above sea level. The soil between the hills is remarkably fertile, and along the Kistna, Godavery, and Wurda rivers and their tributaries is to be found some of the most productive soil of India. Berar is an elevated valley through which flow several large streams that enter into the Godavery and drain a country the soil of which is unsurpassed in richness and depth and adaptability to the cultivation of cotton. From this section comes the cotton known as " Oomrawattee," or Oomras. Dharwar is another good cotton district, being especially suited to the acclimatizing and culture of American cotton. The extent of HISTORY AND GENERAL STATISTICS OF COTTON. 45 territory is small, but being near the sea and possessing a tolerable uniformity of atmospheric moisture, the combination of climate and soil is better adjusted for the production of cotton than any other part of the Deccan, and consequently than any other region of India. Western India is of no special interest in this connection, not being a heavy producer of cotton except in the provinces of Scinde, Cutch, and Guzerat. The soil of these provinces varies in richness and pro- ductiveness from sand to deep black alluvium. The greatest drawback to the cultivation of cotton in this region is the extreme heat and the drought succeeding a rainy season of small precipitation — 3 to 10 inches in Scinde and Cutch, though parts of Guzerat have a yearly fall of 40 inches. Southern India, or the southern part of the Madras Presidency, is best represented in cotton culture by the provinces of Coimbatore and Tinnevelly, which border on the Western Ghauts, where the atmos- phere is humid. The cotton raised in the latter province is the best grown in southern India. Although India has always produced large quantities of cotton, and made most beautiful and delicate webs from its fiber, exporting these fabrics to all parts of the world, it is only within the past one hundred years that she has exported any considerable quantity of raw cotton. The Secretary of the Treasury of the United States in 1836 esti- mated the cotton crop of India for 1801 at 400,000 bales of 400 pounds each; but of this amount, according to Watt and Murray,1 "■the imports into Great Britain from India during the years 1800 to 1809 averaged 12,700 bales per annum," and "in 1818 the export to the United King- dom amounted to 247,000 bales — the largest quantity exported from India up to 1833— but in 1821 had fallen to 20,000 bales." The estimated yield of cotton in India for 1834 was 463,000 bales of 400 pounds each, showing but a slight increase over the production in 1801, and indicating that the increased exportation from that country was at the expense of home consumption, caused by some falling off of the American crop and not an equal increase in production. "The final result of the contest between America and India in cotton supply," says Watt, "is too well known to require any recapitulation. America had gained command of the market, and India was considered only as a supplementary source of supply, resorted to mainly in the event of a short crop in the West." The following table gives the production of cotton in India from I860 to 1894: Production of cotton in India, by periods (in bales of 400 pounds). Tear. • Bales. Year. Bales. 1869 70 1, 985, 000 1891 92 2, 795, 000 1880 81 2,093,000 1892 93 3,225,000 1893 94 2, 902, 000 1890-91 2, 993, 000 1Watt, Dictionary of Economic Products of India, IV, p. 48. 46 THE COTTON PLANT. Important and interesting as are the Indian manufactures of cotton it is impossible in this publication to do more than briefly sketch their progress. The total number of hand looms in the country and the amount of raw material they consume are quite large, but it is impossi- ble to give accurate figures for either. As the power mills of India increased and importation of foreign goods was augmented, the finer fabrics of the hand loom, such as Dacca muslin, once so famous all over the world, ceased to be produced, until now they are rarely seen, and the present generation of weavers, from lack of demand for their prod- uct, have almost lost the art which, transmitted from father to son for 4,000 years, had placed them easily first among the weavers of the world. The first mill built in India was opened for business at Bombay in 1854. This was followed by another in 1855, and by a third in 1857. In 1861 there were 12, with 338,000 spindles and an estimated annual consump- tion of 05,000 bales of 390 pounds to the bale.1 In 1879, twenty-five years after the erection of the first mill, the number had increased to 56, with 1,500,000 spindles. In 1889 there were 124 mills, working 21,600 looms, 2,703,000 spindles, employing 91,600 hands, and consuming 888,700 bales of raw cotton. Three-fourths of this consumption was by mills in Bombay Presidency; Bengal Presidency being second, using 85,000 bales, and Madras Presi- dency third, working up 43,750 bales. In 1893-94 India mills took 1,222,000 bales of 400 pounds net for con- sumption, about 41 per cent of her crop. The amount used outside of the mills was sufficient to bring the total home consumption up to a little over 50 per cent of the crop for that year. India is also a large buyer of cotton yarns and cloth, as in 1892 she imported 50,404,318 pounds of cotton yarn and twist, valued at $6,380,000, and cotton manufactures valued at $52,867,000. India less than twenty-five years ago took 1£ per cent of all cotton grown and now consumes 10 per cent. The growth of cotton manufacture in India since 1861 is shown by the following statistics : Cotton mills of India, [Compiled by tbe secretary of the subcommittee from the annual reports of the Bombay Mill Owners' Association. 1 Tear ending June 30— Num- ber of mills. Number of spindles. Estimated annual con- sumption (iu bales of 392 lbs.). Tear ending June 30— Num- ber of mills. Number of spindles. Estimated annual con- sumption (in bales of 392 lbs.). 1861 12 27 40 47 51 53 56 56 57 65 67 79 338, 000 593, 000 886, 000 1, 100, 112 1, 244, 206 1, 289, 706 1, 452, 794 1, 461, 590 1, 513, 096 1,620,814 1, 790, 388 2, 001, 667 65, 000 114, 000 170, 000 198, 000 215, 000 225, 000 267, 585 307,631 378, 989 397, 562 456, 556 531, 365 1885 87 95 103 114 124 137 134 139 141 142 148 2, 145, 646 2, 261, 561 2,421,290 2, 488, 851 2, 762, 518 3, 274, 000 3, 352, 000 3, 402, 000 3, 576, 000 3, 650, 000 3, 810, 000 596, 749 643, 204 1874 1886 1875 1887 726, 276 1876 1888 786, 982 888, 654 1877 1889 1890 1891 1878 1, 008, 000 1879 1,179,000 1880 1892 1, 166, 000 1, 122, 000 1881 1893 .'. 1882 1894 1, 140, 000 1, 342, 000 1883 1895 1884 About 125,000 persons are now employed in the mills. 1 Watt, Dictionary of Economic Products of India, IV, p. 158. HISTORY AND GENERAL STATISTICS OF COTTON. 47 COTTON IN EGYPT. From time immemorial a fine quality of cotton has been grown in tlie upper regions of the Nile, particularly in Abyssinia. Seed of this variety was brought to Lower Egypt about 1820, and from about this date Egypt has been a regular exporter of cotton to European markets. From 1821 to 1856 the quantity varied from 15,000,000 to 50,000,000 pounds annually; but Egypt felt the stimulus of the cotton famine of 1S61-1865, which caused all cotton-growing countries to increase their yield. In 1800 the export duty of 10 per cent levied by the Government had been reduced to 1 per cent, and this also stimulated the culture, aud the annual average export for the decade 1861-1870 was 310,000 bales of 100 pounds each. Unlike other cot ton- growing countries, Egypt did not reduce its cotton production upon the resumption of shipments from America, but has steadily increased its output, as will be seen from the following table : Exxmrts of cotton from Egypt to Europe and the United Kingdom. Tear. Bales. Tear. Bales. Tear. Bales. 1874 410, 000 347, 000 472. 000 438, 000 400, 000 255, 000 456, 000 405, 000 1882 423, 000 329. 000 384.000 500, 000 410, 000 418,000 411,000 1889 388, 000 433, 000 1875 1883 1884 1885 1886 1890 1876 1891. . 538 000 1877 1892 e^o 000 1878 1893 678, 000 1879 1887 1894 664, 001) 1880 1888 1895 634, 000 1881 According to Foaden1 the present production in Egypt is about 577,500,000 pounds of fiber, practically the whole of which is exported; and 22,275,000 bushels of seed, of which the greater part is exported. Extensive irrigation and drainage systems are in course of construc- tion which will doubtless greatly increase the area of cotton culture. Moreover, other crops are being abandoned to some extent and cotton substituted for them. Mako-Jumel, the name given the variety of cotton first cultivated experienced many changes and evolutions in Egypt, gradually chang- ing its color to a yellowish brown, and this new variety was known as Ashmouni, from the valley of Ashmoun, where this change was first noted. The principal varieties of Egyptian cotton are the Ash- mouni, Mitafifi, Bamia, Abbasi, and Gallini. For many years the Ashmouni formed the bulk of the Egyptian crop, but it is now almost entirely superseded by the Mitafifi. In color it is a lightish brown and its staple is over an inch in length. Its cultivation is continued in some parts of Egypt, but the acreage of this variety is decreasing every year. In Upper Egypt, however, it is more extensively culti- vated, the soil there being less favorable to Mitafifi. 1 Foaden, MS. article on cotton culture in Egypt, iu the possession of this Office. 48 THE COTTON PLANT. The Mitafifi cotton was discovered by a Greek merchant in the vil- lage of that name. The seed has a bluish-green tuft at the extremity, which attracted the merchant's attention, and on planting it he found that it possessed decided advantage over the old Ashmouni. It is more hardy and also yields a greater proportion of lint to the seed. At first from 315 pounds of seed cotton 112 pounds of lint was secured, and sometimes even more. It is now somewhat deteriorated and rarely yields so much, averaging about 106 pounds of lint to 315 of seed cot- ton. The Mitafifi is a richer and darker brown than the Ashmouni. The fiber is long, very strong, and fine to the touch, and is in great demand. In fact it controls the market. Next to Mitafifi, Bamia is perhaps the most extensively cultivated variety in Lower Egypt. It was discovered by a Copt in 1873. The plant is of large size and coarse growth. It is later and less hardy than Mitafifi, and the fiber is poor as compared with that of Mitafifi and Abbasi, light brown in color, and not very strong. In general it may be said that this variety is inferior to Mitafifi in yield, hardiness, and length and strength of fiber. Abbasi is a variety of recent introduction and is not yet very exten- sively grown. It was derived from the Mitafifi through the Zafiri. It resembles Mitafifi, but is somewhat earlier. "The lint is of a beauti- ful white color, fine, silky, very long, though not so strong as Mitafifi, and the first two pickings command the highest price in the market."1 The Gallini cotton, derived from Sea Island and closely resembling it, has almost entirely disappeared from cultivation, as the quality has deteriorated to such an extent that it is difficult to sell. In 1891 only 122 cantars (less than 20 bales) were on the market, and they were not readily sold. Among other varieties may be named Hamouli, of a good mellow brown color, not so long in staple as the Mitafifi, but producing a good I>roportion of lint From 1879 to 1894, on the State domains, the average yield per acre was 425 pounds, valued at $42.59 per acre; from 1890 to 1894, inclusive, the yield of cotton gave on an average $43.14 per acre. During the three years 1892, 1893, and 1894 the money yield, including value of cotton seed, lint, and wood (stubble), has averaged $52 per acre. The cotton seed brings about $4.50 per acre. The wood is valued at about $2.50 per acre, being used as fuel. Foaden gives the following estimate of the present cost of growing an acre of cotton in Egypt: Cost per acre of (/rowing cotton in Egypt. Kent of land, including taxes $27. 00 Irrigation 7. 00 Preparation of land, seeding, manuring, etc 5. 50 Cost of seed 50 Cultivation, including hoeing, thinning, etc 2. 00 Picking 4. 00 Total 46.00 1 Foaden, loc. cit. HISTORY AND GENERAL STATISTICS OF COTTON. 49 The return from lint, seed, etc., is stated by him to be $66, thus yielding a profit of $20 per acre. A small Arab farmer who works his crop with his family can produce cotton much cheaper. The average yield of lint cotton per acre is 340 pounds, but good lands produce 700.1 The cotton gins in use in Egypt are of the roller pattern, there being about 100 ginning mills in the country, distributed in all the chief cotton centers. There are also hydraulic presses attached to nearly all the mills, so that the cotton comes to Alexandria, the shipping point, hydraulically pressed. There are but few steam press mills in the country, so that the great business of preparing for exportation is con- fined to Alexandria. The Egyptian cotton is put up in close, compact, tough covering. The bales, which are long and smooth, weigh from 700 to 780 pounds, occupy from 15 to 20 cubic feet each, are plainly marked, and are wrapped close and bound strong and tight. Being so well covered and bound, injury from fire, water, dirt, and dust is minimized. Ships which can pack 10,000 or 12,000 bales of Egyptian cotton can take only from 6,000 to 8,000 bales of American cotton, although, according to the ratio of weights, they should take 14,000 bales. As it is, the cost of freighting is much heavier for American cotton per pound than for the Egyptian, and there is much waste in unnecessary packing material, and loss by dirt, mud, and bursting of bales, which affect about equally the producer and the manufacturer. Egyptian cotton is not as fine as Sea Island cotton, and of course does not bring so high a price, but is better than American upland cotton for goods requiring smooth finish and high luster. It gives to fabrics a soft finish somewhat like silk. It has a strong, silky staple from 1£ to If inches in length. The imports of Egyptian cotton into the United States for the sea- sons ending August 31 were at different periods as follows: Imports of Egyptian cotton into the United States. Tear. Bales. Tear. Bales. 1884-85 4,553 3,815 4,700 5,792 8,430 10, 470 1890 91 23, 790 1885-86 1891 92 27, 739 1886 87 1892 93 42, 475 1887-88 1893 94 33, 606 1888-89 1894 95 59, 418 1889-90 | COTTON IN BRAZIL.2 The consensus of the early historians of Brazil is that cotton is indigenous to that country, and that while the natives probably did not 'See also article on culture of cotton. 2 The principal authority upon which this account rests is Branner, Cotton in the Empire of Brazil, U. S. Dept. Agr., Special Report No. 8, 1885. 1993— No. 33 4 50 THE COTTON PLANT. plant it they made use of the fruit of the wild plant to supply their few and simple needs for clothing and other textile products. Beauchamp, l treating of the period immediately following the dis- covery (1500 to 1521), refers to cotton cords used by the Indians of Brazil upon their bows and for other purposes. Visconde de Porto Seguro,2 in describing the customs and utensils of the ancient inhabit- ants of the Upper Amazon Valley, says : "They made use of blowguns, the arrows of which were wrapped with cotton." Auguste de Saint- Hilaire,3 one of the most trustworthy authorities and careful observers who have traveled in Brazil, says that the earliest travelers found cot- ton in use among Indians along the Brazilian coast; that they used it for making cords, hammocks, and even clothing, giving as one author- ity for this statement Hans Stade, who was held in captivity in south- eastern Brazil from 1547 to 1555. Abundant evidence of the fact that cotton was used by the Indians at the time of the discovery of the country is furnished by the oldest documents upon Brazil referring to that event, and by the letters of the Jesuit missionaries who immedi- ately undertook to Christianize the Indians. Also the existence among the aborigines of words signifying cotton points to their knowledge of cotton, and very probably to some of its uses. Nearly all of these words appear to be the same, or at least have the same origin. The slight differences that exist are probably caused by the differences in the dialects spoken by the various tribes of the Indians, or may be attributable to the differences among the observers and the different nationalities of the observers. One remarkable thing about these terms is that none of them bear any resemblance to the European words for cotton. Gabriel Soares de Souza, who lived in what is now the province of Bahia from 1570 to 1587, describes in detail some of the Indians then inhabiting that part of Brazil. " It is customary for the men,'f he says,4 "to wear their hair so long that it reaches their waist, and sometimes they have it braided and intertwined with strips of cotton, so that it looks like a broad braid. The women are close shaven and wear about themselves aprons made of cotton thread, with a long fringe." Claude d' Abbeville, a Capuchin missionary in Marauhao from 1612 to 1614, also reports the natives of that country as using cotton ham- mocks. In another place he says:5 They gather, clean, beat, and spin cotton with much dexterity, and with it make open hammocks resembling nets, and others as well woven and full of figures as if they were the work of better weavers; also aprons in which they carry their chil- dren about their necks. 1 Histoire du Bre"sil, Vol. I, pp. 93, 96. '2Historia Geral du Brazil, Vol. I, p. 30. 3 Voyage daus le District des Diamans, pp. 253, 254. 4Revista do Instituto Historico do Brazil, p. 352. 6Historia do Marauhao, p. 356. HISTORY AND GENERAL STATISTICS OF COTTON. 51 In 1570 to 1587 De Souza wrote of the cotton plant in Bahia: "These cotton plants are cleaned with the hoe two or three times during the year to keep the weeds from choking them." Notwithstanding the general cultivation of cotton throughout Brazil, there is no record of its having been exported to any great extent, and but little evidence of its being exported at all, until the middle of the seventeenth century. Marques1 says that after the formation of the commercial company of Maranhao and Gram-Para the first exportation of cotton took place in 1760, which consisted of 651 arrobas, or 20,831 pounds. Camara says, in a memorial upon the cultivation of cotton published in 1719 in Lis- bon, that cotton was first sent from Pernambuco to Portugal in* 1778, and that up to 1781 the quantity shipped was very small. J. B. Lyman2 says that the export of cotton from Brazil to England began in 1781. A commission 3 appointed by the Government in 1852 to revise the tariff made the following report regarding the early exportation of cotton : In some parts of Brazil in remote times there was no exportation of cotton — the cultivation was limited to what was necessary for use in that country. It was from Parahyba that it was first exported, whence it was sent to Portugal. Pernambuco was given up for a long time to the production of sugar, aud it was ouly in 1778 that it exported this article for the first time. Its exportation, however, was very small until 1781. Some side light is thrown on this subject in the story of the ship- wreck of Jorge de Albuquerque Coelho.4 His vessel left Pernambuco on the 16th of May, 1565. On account of the condition of the vessel, they put back to port, and left again on the 29th of June, 1565. Their passage was a stormy one, and the sea became so rough at one time that they were obliged to throw over part of their cargo. The following is an extract from their report: "And seeing that all this was of no avail, and the winds grew higher, as if they wished to overwhelm us, we threw overboard the artillery, many boxes of sugar, and many bales of cotton." Now, this vessel was loaded in the province of Pernambuco, and the many boxes of sugar and bales of cotton must have been grown by the Portuguese within this province. By the end of the seventeenth century the cultivation of cotton had become general throughout almost the whole of Brazil, aud consid- erable quantities were being exported to Europe. During the eight- eenth century the more general use of gold as a circulating medium, the removal of prohibitory laws, and the increased demand for raw 'Marques, Dictionary of the Province of Maranhao, p. 13. -J. B. Lyman, Cotton Culture, p. 153. 3Relatorio da Commissao da Revisao da Tarifa ao Governo Imperial, 1853. 4Revista do Institute Historico do Brazil, p. 279 et seq. 52 THE COTTON PLANT. material in Europe led to what was for those times an extensive culti- vation and exportation. The only statistics to be obtained up to the end of the eighteenth century are those of the province of Maranhao. From these we find that in 1800 5,529,408 pounds were exported, but this port stood only second among the cotton-exporting ports. Pernambuco probably ex- ported twice as much, while Bahia, Rio de Janeiro, and Para together exported as much as Maranhao. Visconde de Porto Seguro says that 70,000 bags of cotton, 1G5 pounds to the bag, were exported, of which 40,000 were from Pernambuco, 16,000 from Maranhao, 10,000 from Bahia, and 4,000 from Para and Rio; and in another place he says that Ceara exported 40,000 sacks. Upon the arrival of representatives of the royal family of Portugal (April, 1808) Brazil ceased to be a mere colony, and a new impetus was given to this as well as to the other industries. Ports were made free to friendly foreign powers, and the decree prohibiting the use of looms for other than the coarsest kinds of cotton was revoked. Cotton had now become a regular and constantly increasing article of exportation, being in such demand and commanding such prices that it was brought from great distances inland on the backs of mules and horses and over almost impassable roads. Such was the condition of the cotton industry in Brazil at the time the United States entered the market as a cotton producer, and the effect of her rivalry in one of the most important branches of agri- culture was very disastrous to cotton production in Brazil. The territory of Brazil capable of yielding cotton is coextensive with the Empire itself. Cotton grows in almost every one of the provinces, and in regard to the others there exists little doubt of its adaptability to this plant. Various authors mention cotton as grown on the Tapajos and the Madeira rivers. From Sao Paulo all along the coast to the Amazon, and for that matter throughout the whole Empire, cotton may be grown in almost unlimited quantities. In reality, however, its cultivation to a considerable extent is limited to the drier regions of the north, along the valley of the River Sao Francisco, and in some parts of the prov- ince of Minas Geraes. In the north — that is, to the north of Sergipe — a belt about 50 miles wide along the coast is for the most x^art devoted to sugar. Immediately beyond this is the region in which cotton is actually grown. The width of this cotton belt is restricted by inade- quate transportation facilities. Transportation is principally on the backs of horses, and the railways are as yet too few to influence the amount of cotton produced. As a rule, it may be said that the cotton belt in the north begins at the edge of the sugar belt, 50 miles inland, and extends about 200 miles inland, more or less, to the south ; as it approaches the province of Bahia it extends inland, and keeps mainly within the region drained HISTORY AND GENERAL STATISTICS OF COTTON. 53 by the Eio Sao Francisco. In the more southern provinces, from Espi- rito Santo and Eio de Janeiro, the amount of cotton produced at pres- ent for exportation is insignificant. Formerly, especially during- the prevalence of the high prices caused by the civil war in the United States, cotton was grown in some of these southern provinces for exportation, especially Eio de Janeiro and Sao Paulo. The island of Fernando de Noronha is an exception to the statement that the cotton is all grown inland. During the civil war in the United States and for some time afterwards the finest cotton sent from any part of Brazil was grown on this island, and was comparable with the Sea Island cotton, possibly fully its equal. But though efforts are being made to renew the cotton culture on the island, which went into decline in 1870 or thereabouts, the success has not been very great. The soil is remarkably fertile and the island is one of the best culti- vated pieces of soil in Brazil. It must not be supposed that only native species are grown, or, indeed, that any of the kinds commonly cultivated are native, though it is pos- sible that one of the species supposed by some authors to be exotic is indigenous to South America. Planters seem to know of but three kinds, and even in these the differences are not always persistent. As a rule, the varieties are larger than ordinary cotton, and some of them are perennial. None of these varieties of cotton produce more than one crop per annum, many assertions to the contrary notwithstanding. The even temperature throughout the year and the favorable weather lengthen out the picking time to several weeks and even months, and it is possi- ble that those who state that there are two or more crops in a year mis- take the various pickings for successive crops. Sea Island cotton has been introduced in Brazil, but without success. The complaint that Brazilian cotton is dirty is not due to any quality of the cotton, but, like the Indian cotton, to the manner in which it is picked, cleaned, and handled. In the Eoteiro do Brazil a writer puts in a few words the process of cultivation in use in Bahia in its early history, and probably in all Brazil, as follows : These cotton trees last seven and eight years and more if the ends of the large branches are broken off (for they dry up), in order that they put forth other new and more vigorous ones. These cotton plants are cleared with the hoe two or three times a year to keep the grass from crowding them. In the preparation of the soil "all the planter has to do is to burn off the woods and plant his seed at the proper season." This is the whole story. There is no uprooting of stumps, no picking out of sprouts, no breaking up with a plow, no preparation of soil, no laying out of furrow, no cultivation- other than the occasional chopping out with the hoe of weeds or sprouts. Therefore it is evident that the cultivation of cotton is almost without labor; in fact, Auguste de Saiut-Hilaire says: 54 THE COTTON PLANT. "iNotliing in this country is less expeusive or more productive than cotton culture." In some sections the saw gin is used, though it has been found by experience to injure greatly the fiber of the Brazilian staple, and the common hand-roller gin is more frequently used. In connection with the gins there are rude presses, where the cotton is made into bales of about 385 pounds. A hand screw is the power used in this baling machine. Generally the seed are not turned to account. In a few instances, near the seaport, they have been shipped to England, or where steam engines are used for baling or ginning they are some- times used as fuel. The greater part, however, is left to rot upon the ground or to be eaten by the cattle in the neighborhood. Only occa- sionally are they used as manure. The home consumption of cotton is very large and is increasing. This is because of the difficulty of getting the raw material to market from remote points, the evenness of the temperature, which does not require warm clothing, and the high tariff upon foreign manufactured goods. Much of this manufacturing is done in the homes of planters, there being comparatively few manufacturing establishments in the country. These establishments are mostly in the provinces of Eio de Janeiro, Minas Geraes, Sao Paulo, and Bahia, where the demands for ordinary grades of coarse cotton cloth are greatest, but they by no means have done away with domestic consumption of the raw material. There is no more familiar sight to the traveler in the interior of Brazil than that of spinning with the ancient distaff and spindle. Brazil undoubtedly weaves many thousand pounds of her cotton annually in that way. , The amount of cotton produced in Brazil in 1859-60 was 70,000 bales; in 1893-91, 300,000 bales. The amount manufactured in 1859-60 was 10,000 bales; in 1893-94, 100,000 bales. Brazil exports about 150,000 bales of 400 pounds each to Europe, most of which goes directly to England; and should there be a demand sufficient to justify the planters to increase their acreage, the crop could be increased many- fold — in fact, there is sufficient good and available land to produce 40,000,000 bales, at a yield of only 100 pounds of lint per acre. As population increases and railroads are extended the cotton crop of Brazil may be expected to increase commensurately. COTTON IN RUSSIA. The Russian cotton-raising district lies in her Asiatic territory, in Turkestan and Transcaucasia. Russian Turkestan is bounded on the west by the Caspian Sea, Ural River and Mountains, on the east by the Pamir Plateau, Tian-Shan and Altai ranges, on the north by the Kirghiz steppes, and on the south by Afghanistan and Persia. West- ern Turkestan is commonly supposed to consist of vast low-lying sandy plains, but nothing could be more opposed to the actual conditions, for HISTORY AND GENERAL STATISTICS OF COTTON. 55 the relief of the laud here presents greater contrasts than are else- where found on the surface of the globe. The lowlands bordering the Caspian Sea are sometimes as much as SO feet below the sea level, while the highlands of the Tian-Shan peaks rise 25,000 feet above the sea. The Aralo-Caspian basin is watered by the rivers Amu-Daria, Zerafshan, Murghab, and Syr-Daria. The low-lying and level parts of Turkestan are covered with marine and lacustrine deposits of geological epochs preceding the present one. Here are found conglomerates, sands, and porous or loose soils. These recent deposits are well adapted for cotton growing. In this district cotton has been cultivated from most ancient times. In all probability this culture began at the time when the Asiatic cot- ton plant (Gossypium herbaceum) began to develop on the borders of the Oxus (Amu-Daria) and Seikhoun (Syr-Daria), to which it had been brought from southern Asia. After the plant became acclimated in that northern latitude several new varieties were produced, and its cul- tivation contributed much to the welfare of the inhabitants, who pro- duced sufficient for their own use and exported some to neighboring countries. During this time there were no planters who devoted them- selves entirely to the cultivation of cotton. It was grown on lands unoccupied by such prime necessities of life as wheat, rice, barley, and other staples. Such was the state of cotton growing in Turkestan before the Rus- sians entered that territory. The quantity produced fluctuated accord- ing to the prices and demands from other countries, Russia being the largest customer. Cotton cultivation attained its greatest development soon after 1860, when the Russian cotton trade, together with that of other European nations, was undergoing a severe crisis in consequence of the great decrease of fiber from the United States. The prices being very high at that time, its production increased in the Bokhara, Khiva, and Khanstvo districts, and also in the Transcaucasian dis- tricts, to such degree that the crisis passed more or less fortunately. After the Russians had thoroughly conquered Turkestan the quantity of cotton grown in that province decreased very rapidly, chiefly on account of the decline in prices. Russia, however, soon found the value of this product to her textile industries and began to devote a great deal of attention to the development of the plant. As it was of great importance, not only to the province itself, but also to Russia, which had been obliged to buy all raw cotton from the United States and Egypt, the Government gave the culture every encouragement. As in all other cotton-raising countries, an effort was made to intro- duce the American cotton into Turkestan, but all attempts to cul- tivate the imported seed were for ten years very unsuccessful, mainly because Sea Island was most frequently planted and the dry weather of Turkestan was unsuitable to this variety. In 1880 it was discovered that upland cotton could be successfully grown there, and energetic 56 THE COTTON PLANT. measures were taken for its introduction and cultivation. The Govern- ment established a cotton plantation at Tashkend, manuals for the cultivation of the American upland cotton were published in the Rus- sian and local languages, seeds were distributed free of cost to those who desired them, and the sale of the cotton fiber raised from the imported American seed was guaranteed. Methods of ginning and cleaning the cotton were also improved, gins being ordered from the United States, with the result that cotton fiber of the American variety grown in Turkestan had a great influence on the Russian market and its price was much better than that of Asiatic varieties. The construction of the Transcaspian railway also had an important bearing on the increase of cotton growing in this country, and the area under crop increased from 810 acres in 18S4 to 120,150 acres in 1889. In 1890 there were 245,000 acres planted in cotton in Turkestan, from which more than 45,600,000 pounds of clean fiber were collected, an average yield of 180 pounds per acre. Of the three districts — Syr-Daria, Fergana, and Samarkand — in which most of this cotton is raised, Fergana is much the largest producer. While the production of cotton in Turkestan will in all probability increase still further, it must be borne in mind that the increase is limited by the extent of the artificially irrigated lands suited to cotton growing; yet within this limit, by the rational employment of the water supplies, especially those of the Syr-Daria, and the construction of new irrigation systems, the quality of the crop may be improved and the quantity largely increased. The methods of cultivation of the cotton plant in this region are exceedingly varied and have not been definitely fixed. The more intelli- gent growers are still experimenting and seeking new methods. Differ- ent systems of plowing and preparation of the soil, of sowing and irrigating, are being adopted in the hope of thus obtaining better results. Yet, on the whole, the methods of cultivation have changed compara- tively little and are being very slowly perfected. The preparation of the fields with improved implements, the manuring of the soil, and the careful selection of seed are only found on the few Russian growers' plantations around Tashkend, and still more rarely in the territories of Fergana and Samarkand. " The local planter breaks up the soil with the aid of a primitive wooden plow known as the csokha' (a turn plow is unknown), harrows the field with a single board, covers the seed by hand, applies manure only in exceptional cases, and pays no attention to the choice of seed, which he sows broadcast" — in fact, makes use of the methods inherited from past generations of ignorant cotton raisers — yet, thanks to the exceedingly favorable conditions of soil and climate, generally harvests a good crop. Most of the plantations are small, those of over 270 acres being exceptions. These belong to Russian planters and trading firms. The large majority of the holdings consist of plats of from 1 to 15 acres and HISTORY AND GENERAL STATISTICS OF COTTON. 57 are cultivated by tbe natives, yet these small cultivators supply at least 90 per ceut of the total production. The average crop of upland cotton in the Syr-Daria territory can be taken as 25 pounds of pure fiber per acre, while that of the local variety is about 180 pounds. In the Fergana and Samarkand territories the average yields from the same area are 250 pounds of upland and 100 pounds of local. Of course, in some cases, particularly when the autumn is warm and free from frost, considerably heavier crops are secured. There are instances of yields of 150 pounds of pure upland and 360 pounds native cotton per acre. Almost all the native cotton is still cleaned by wooden machines worked by hand power, which while not rapid are very cheap. Nearly all the upland cotton is sent to special, cleaning mills, where improved gins, generally worked by water power, though sometimes by animal power and less frequently by steam, are to be found. These gins are mainly in the towns which serve as centers of the cotton-growing dis- tricts, but a few are to be found in the larger settlements around which are extensive plantations. In 1893 there were about 100 mills in all Turkestan, in which more than 400 gins and 120 presses were at work. Most of these gins were imported from the United States, but there were also Russian-made gins. The seed is used for planting, for the production of oil, and for fuel. The cake left over from the oil press is utilized as cattle food. The lint is pressed into bales of from 250 to 325 pounds, which are then dis- patched to Samarkand, the terminus of tbe Transcaspian railway, upon the backs of camels or in a rude cart known as the "arba." The camel load is 2 bales, or 500 pounds, and an arba load G bales, or 1,500 pounds. The almost impassable condition of some of the roads makes the transportation of cotton very slow, and a periodical delivery at the railroad station is not yet to be thought of, but when new lines of rail- way which have been projected into the cotton-growing districts are finished, especially that from Samarkand to Kokand, penetrating the Fergana territory, transportation will be improved and the cost of delivery to the railway, and consequently to European Russia, will be cheapened, thus increasing the demand and inducing larger plantings. Cotton for European Russia is shipped from other countries of cen- tral Asia, namely, Bokhara, Khiva, and the Transcaspian territory. Bokhara produces about 51,000,000 pounds of cotton, mostly of the Asiatic variety. This cotton is shipped by the Transcaspian line of rail- way. Khiva produces about 21,000,000 pounds, almost exclusively a native variety, but considerably superior in its quality to the other Asiatic growths. The larger portion of this cotton enters Russia through Orenburg by camel caravans; the rest is dispatched by boats by the way of the Amu-Daria to a station of that name on the Transcaspian railway, and thence by rail. 58 THE COTTON PLANT. The Transcaspian territory produces but 360,000 pounds of cotton, mainly the upland variety, and most of that in the neighborhood of Merv. Thus all the central Asiatic countries together produce 144,000,000 pounds of cotton, more than three-fourths of which is sent to European Eussia. Another considerable district of cotton culture in the Eussian Domin- ion is Transcaucasia, where upland cotton has made little progress in displacing the variety cultivated by the natives from very early times. There are about 100,000 acres devoted to cotton in Transcaucasia, the greater part of which is in the Erivan government. The total produc- tion in 1891 was about 22,000,000 pounds, the average crop being 230 pounds per acre of upland, the local plant not yielding quite so much. The methods of cultivation are. being perfected and the quantity of cotton produced is increasing. ;- It may be said in general that the raising of American cotton has been making considerable progress in Eussia and Turkestan. In the year 1892 1,080 acres were planted with American cotton in the district of Dshisak, Province of Samarkand, producing a crop of 783,000 pounds of raw cotton, from which 234,000 pounds of clean cotton was obtained — that is, 216 pounds per acre on the average. Eeports of a later date are not available. Experiments on methods of culture and test of varieties are still being carried on in this region under the supervision of the Eussian Government. The American upland variety known as Ozier Silk is generally grown. Sea Island cotton will not mature in this region which is near the northern limit of successful cotton culture, but it has been found that with upland varieties the season is shorter and harvesting is usually finished by the time that it begins in the southern United States. The quality of tbe fiber is apparently fully equal to that of the American product from the same varieties. These results are not surprising for it is a well-known fact that " nowhere are plants cultivated so advantageously, whether from the point of view of quality and quantity or that of cheapness of production, as at the northern limits of their sphere of cultivation." In the whole Province of Samarkand there are 8 cotton gins; there were in 1890 5,764 acres planted with American cotton and 15,762 acres with domestic cotton; in 1892 the proportion was reversed, there being 12,204 acres devoted to American cotton and 3,510 acres to domestic cotton. In the Province of Khojend American cotton is rapidly taking the lead, and the cotton acreage is also increasing; in 1892 there were pro- duced 3,780,000 pounds, which equaled 244 pounds of cleaned cotton to the acre on a general average. A steady progress of cotton culture would appear from the statement that the quantities of cotton carried by the Transcaspian railroad to the Eussian market have increased from 31,428,000 pounds in 1888 to 129,168,000 pounds in 1893, and that of the whole bulk of cotton HISTORY AND GENERAL STATISTICS OF COTTON. 59 shipped in 1894 72,000,000 pounds was American cotton, while in 1884 there were only 250,000 pounds of that variety raised. It is said that the plant deteriorates in Turkestan and can only be kept up to its high standard by frequent renewal of American seed. Notwithstanding the large production of cotton in this portion of the Russian Empire and the fairly rapid development and increase of the crop, the time is probably still far distant when Russian cotton mills will manufacture exclusively Russian-raised cotton. The supply of cotton from central Asia in the Russian markets, though largely increased of late, is still far from being sufficient to satisfy the con- tinually increasing demand, and the quantity of cotton imported is diminishing but slowly. The amount of cotton brought into Russia from abroad in 1891 was 351,939,412 pounds, the average for the ten years preceding being 290,000,000 pounds. The greatest quantity of cotton used in Russia comes via the European frontier from the United States, Great Britain, Germany, and Egypt. Great Britain and Ger- many send American, Brazilian, Indian, and perhaps also Egyptian cotton. In 1891 Russia imported from the United States 136,749,492 pounds, from Great Britain 23,597,820 pounds, from Egypt 04,749,492 pounds, and from Germany 24,347,73G pounds, and used about 108,- 000,000 pounds (nearly one-third of the total consumption) of the prod- uct of her Asiatic dominions. The entire consumption of cotton in European Russia appears to have been 500,000,000 pounds in 1892 and about 70,000,000 less in 1893. It is possible that the construction of railroads into the cotton-growing regions and improvements in methods of culture will ultimately result in a balance between Russian exports and imports of cotton. This view appears to have some confirmation from the figures of the cotton trade of the United States with Russia. During the four years 1881-1884 the United States exported annually an average of 124,117,441 pounds of cotton to Russia. In the period 1885-1888 the average was 86,014,804 pounds, and it fell to 75,889,614 pounds in the period of 1889-1892. The decline was from 134,000,000 pounds in 1881 to 67,000,000 pounds in 1892, a fall of 50 per cent. Thus, while there is probably no immediate prospect of Russia producing all the cotton she needs, our own declining trade gives some warrant for apprehension that in the future our cotton planters will meet a competition from this source. COTTON IN JAPAN. Cotton (lint, wata; plant, Ica-tvata) is cultivated to a considerable extent in Japan, where it was introduced from India many years prior to its introduction into China. It has ordinarily been said that the introduction of cotton was from China in the sixteenth century, while in fact the cotton plant was cultivated to a limited extent in Japan several hundred years previous to that date. Fesca1 says that cotton 1 Japaiiiacken Laiulwirtlischaft, 1893, Pt. II, p. 48f>. 60 THE COTTON PLANT. was first accidentally introduced into Japan in the year 781 A. D. from India, and that its culture soon ceased; that again, several centuries later — probably 1592 — it was introduced by the Portuguese, though this is not positively authenticated. However, it was in the seventeenth century, during the reign of Tokugawa, that the spread of cotton cul- ture in Japan began, which has continued until the present time. According to a paper by Poate read before the Asiatic Society of Japan in 1870, one of the old .and sacred books of Japan, called Wajishi, contains an account of the introduction of cotton into that country which supports the above statement. The cleaning of the cotton is largely done with the scutch, or bow, and the roller gin, or churcka, is in common use. The lint is ordinarily packed in 100-pound bales. The quality of the Japanese cotton — at least that produced in the best cotton districts — is good, though the staple is considerably shorter than the ordinary tree or shrub cotton. The staple is seldom more than 1.8 centimeters in length. The best cotton is also elastic and glossy, but the gloss appears to be more a result of climate and locality than of variety, a damp climate being unfavorable to the production of glossy fiber. The cotton of the damp San-iudo region has little gloss, while that of the neighboring province, especially that of the Osaka Fu and Wakayama Ken, possesses a very good gloss. The last-named lint is considered the best. It is of medium length, soft, very elastic, and glossy as silk. The strength of the lint in the province of Suo is well known. That produced in the San-indo district is mostly short staple, hard, and without gloss, probably on account of the low temperature aud humidity of the climate. In the provinces of Kuantoebene and Hitachi, and on the north edge of the cotton region, there is produced a long and fine staple cotton which is quite elastic and strong. The relation of lint to seed is very favorable. Generally it stands GO per cent seed and 40 per cent lint. The cotton of Kuantoebene is an excep- tion to this, being only between 20 and 30 per cent lint.1 Statistics of the total product of the Empire are to be had for the years 1878-1884, 1887, 1891 in unginned cotton and are as follows: Total production of unginned cotton in Japan."1 Tear. Pounds. Tear. Pounds. 1878 118, 958, 541 207, 888, 208 118, 685, 333 120, 678, 880 115, 162, 850 1883 139, 510, 675 1879 1884 127, 493, 691 1880 1887 186, 571, 583 1881 1891 109, 839, 383 1882 The acreage under cotton, while not in itself large, is in fact much larger than that devoted to any other textile fabric, and in 1887 was 2 M Japaniscken Landwirthschaft, 1893, Pt. II, pp. 474, 475. "The present production of ginned cotton is given as about 75,000 bales (of 400 pounds) by Shepperson. HISTORY AND GENERAL STATISTICS OF COTTON. 61 per cent of the entire cultivated area. In west Japan, which produced large quantities of cotton, the cotton fields occupy 4.8 per cent of the entire arable land, against 2.77 per cent in middle Japan, 1.45 per cent in Shikoku, 0.45 per cent in Kiushiu, and 0.34 per cent in north Japan. West Japan furnishes nearly half of the total production, 49.48 per cent; middle Japan, 40.07 per cent, although the cotton area of west Japan is only 80 per cent of that of middle Japan. During 1891 the amount of cotton raised. was 109,000,000 pounds, and in 1892 Japan imported 100,000,000 pounds of raw cotton, of which not more than one-tenth came from the United States, two-fifths from China, and one-half from all the other countries; and her exports amounted to 325,000 pounds of raw cotton, of which 300,000 pounds went to Korea, leaving for home consumption somewhat over 208,000,000 pounds. During the years 1891 and 1892 the average value of Japanese coins varied so much that though the total value of trade for 1892 exceeded that of the previous year by 19,974,000 yens, yet when reduced to United States money at the rate of exchange for that year the increase was only $761,000, making it difficult to satisfactorily compare amounts with previous years. The most important thing concerning the foreign trade of Japan is the increasing facilities of the country for manufac- turing her own supply. Especially is this noticeable in relation to cot- ton goods. New spinning mills are being established from time to time, and large profits are made from those already at work. The consump- tion of cotton goods is increasing each year, and home factories are supplying a larger aud larger percentage of the entire amount con- sumed. In 1892 Japan imported $9,250,000 worth of textile fabrics, of which the larger portion was probably cotton, besides $8,750,000 worth of raw cotton and $5,000,000 worth of cotton threads and yarns. COTTON IN CHINA AND KOREA. The Chinese began to make extensive use of cotton fiber about 1300 A. D., and since that date the growth and manufacture of cotton has reached immense proportions in the Celestial Empire; but statistics of the cotton crop in that country are not attainable, the closest estimate being 1,G00,000 bales of 400 pounds each as a product of China and its former dependency Korea at the present time. The great cotton region of China lies along and on both sides of the river Yang-tze-Kiang, where the soil is very fertile, x>erhaps surpassing any other district of China. The method of cultivation is very primitive. In some cases the ground is plowed in March, or as soon as the frost is out of the ground, and the seed is sown broadcast. Under this system of cultivation the plants grow very thickly, almost as close as cereals; they are spindling and short and the bolls are small. The yield per acre varies from 50 pounds to 1,000 pounds of unginned cotton. The method of cultiva- tion varies, however, in different parts of China, for in other cases after 62 THE COTTON PLANT. plowing the seed is planted in hills on ridges and cultivated with the hoe. About five months after planting picking commences, this being the work of women and children ; after the crop is gathered the stalks are pulled up and preserved for fuel. There are no large cotton plantations in China, most of the cotton being raised on patches varying in size from a few square yards to one or two acres. The old Hindoo churcka is generally used for ginning, the scutch or bow for cleaning, and the spindle and handloom for manu- facturing, but cotton factories are being established in some localities. • The quantity consumed can not be determined from any statistics kept at the open ports. The farmers do not usually sell their cotton, but it is spun and woven by the women, and any surplus cloth is sold in the neighborhood. They make use of the seed for the extraction of oil, and use the oil for illuminating purposes. China exports some cotton to Japan, but the imports of yarn and cloth from Japan, India, and Eng- land are far in excess of the exportation of raw cotton. Though the books of the treasury of Korea show only 190,000 acres devoted to cotton culture, it is probable that owing to the system of taxation the estimate is not more than one- fourth of what it should be, and that the total acreage is about 800,000 acres. On this is produced annually about 800,000,000 pounds of unginned cotton. It is cleaned by hand machines and yields about 200,000,000 pounds of lint, or about 250 pounds per acre. Most of this is consumed at home, giving about 13£ pounds for each person. It is grown chiefly in the provinces of Huang- Hai, Chel-La, and Kyeng-Sang, and to some extent in Chung-Cheng and Kyeng-Kwi. Because of lack of information on Korean affairs, it is quite impossible to say how much of the land is suitable for cotton, but it may be said that the area under culture is practically dependent upon the price of the fiber. Probably one-half of the tillable land might be so cultivated, although only a small portion is now used. Cotton was brought to Korea about 500 years ago from China. The soil and climate have improved the original plant, giving it a longer staple and finer fiber. It is the perennial variety, but it is found more profitable to uproot it and replant each year. The stalk is used for fuel and its ashes for manure. Chinese usages are closely followed. Korean cotton fiber is considerably better than that of China or Japan in dura- bility and warmth-retaining qualities. Cloth is all made in nrivate families and for private use, but no statistics can be obtained. COTTON IN OTHER COUNTRIES. At least nine-tenths of the world's supply of cotton is produced by the countries already referred to, but there are several other countries which either already produce small quantities of cotton or might enter HISTORY AND GENERAL STATISTICS OF COTTON. 63 the list of cotton-producing countries. They have climate and soil suitable to the growth of this plant, but are prevented at present from entering this industry upon any extended scale either because of scanty population, difficulty of transportation, or greater profit gained from the cultivation of other crops. The total amount produced by these countries as a whole iu 1893 is estimated at about 130,000 bales. Of these, Mexico, by reason of its proximity to the United States, is probably for us the most important. It has been estimated by Ruiz tbat the annual production of cotton in Mexico prior to the conquest by Cortez was about 116,000,000 pounds, but under the rule of the Spaniards the cultivation declined until it was entirely abandoned in many sections of the country. Some new impulse was felt about 1860, at which time every country that could produce cotton was called upon to attempt to make good the decrease of receipts by European countries from the United States. Since 1882 a larger interest has developed in Mexico in the culture of cotton, so that every State suitable for its culture has more or less area devoted to the plant. Ruiz, in his report on " Cotton in Mexico," estimated that in 1892 the output equaled 25,000,000 pounds of ginned cotton. The greater part of the Mexican crop is produced in the State of Coahuila, but there are three well-defined cotton sections in Mexico — one along the eastern coast, one in the central plateau, and the other on the west coast, of which the inland section is the largest producer. The crop there, however, is dependent upon the water available for irrigation. Biaconi1 gives it as his opinion that if all the land suitable for the purpose were put under cultivation the Mexican Republic might easily become a rival of the United States in the production of cotton. As the conditions now are, Mexico is obliged to purchase cotton for its own needs from the United States. The best cotton is produced in the State of Guerrero in the neighborhood of Acapulco, and the most inferior is produced in the State of Chiapas. In the zone along the Gulf of Mexico, 300 miles long by about 50 miles wide, cotton could be raised, but a lack of labor is a great drawback to this enterprise. Besides, coffee growing offers so much greater profit that few agricul- turists give serious thought to tbe growing of cotton. On the Pacific Slope the cultivation of cotton comprises the whole coast, where there is much laud that exhibits great fertility; on account of the cost of trans- portation and lack of railway facilities, however, the extension of cotton culture has been retarded. Although in the interior of the country there exists no continuous stretch of land especially favorable to the culture of cotton, there are, nevertheless, centers which are capable of becoming large producers of the staple. The State of Coahuila is situated in this section, and Laguna is the most important of the cotton- 1 "Le Mexico." 64 THE COTTON PLANT. growing districts of the State, as it produces at least one-half the cotton raised in Mexico. It is about 360 miles southwest from Eagle Pass, Tex., and is skirted by the Mexican International and Mexican Central rail- ways, being therefore provided with an outlet for its cotton crop. Cotton manufacturing — at least as the term is understood in these modern days — has received an impulse only within the last few years, but is extending itself so that there is hardly a State that has not an establishment for spinning and weaving cotton, and these factories not only consume the entire home product, but also from 50,000 to 70,000 bales of cotton from the United States, the amount increasing as the shortage in the home-raised crop necessitates larger imports from abroad. PERU. Cotton is indigenous to Peru, or at least it has grown there from prehistoric times.1 The inhabitants at the time of the conquest by the Spanish troops were clothed in cotton, and mummies of very ancient date have been found wrapped in cloths or blankets wholly or in part composed of cotton, but, as was tbe case in Mexico, the culture of cot- ton was much neglected by the Spanish conquerors, whose eager search for precious metal led them to ignore the greater agricultural wealth of their colonies. Statistics as to the output of cotton in Peru are not available, nor are the exports of cotton to other countries to be had for dates earlier than 18G2, when 341,243 pounds were exported to Liver- pool. In 1865 the exports had increased to 4,145,260 pounds. From 1885 to 1892 the average export of cotton from Peru to Liverpool was about 6,000,000 pounds, to which must be added the amounts shipped to New York. The variety Gossypium barbadense peruvianum, with a rough, strong fiber, is cultivated along the banks of rivers and lowlands irrigated by the overflowing streams. It is perennial, and when con- ditions are favorable will bear two crops a year for many years, but the ordinary productive stage is about seven years. The rainy season ends about the middle of April, and the first crop is gathered in February. The second crop is probably the largest of the series, with a gradual diminution with each successive crop until the yield is no longer remu- nerative. The chief use of this cotton, which is known as "Bough Peruvian" in commerce, is for mixing with wool. It lessens the tend- ency of the goods in which it is used to shrink, makes them more dura- ble, and gives a better luster and finish; hence it is frequently used in the manufacture of underwear and hosiery. This peculiarity of the Peruvian cotton is probably the result of soil and climate, and its cul- tivation is therefore likely to be restricted to that country. It would be very difficult to find a section in the United States that would fur- nish a uniform and high heat during the ten months necessary for the development of the plant, or the other conditions which contribute to the successful cultivation of this cotton. 1 Auxiliador da Industria Nacional, 1878, p. 90. HISTORY AND GENERAL STATISTICS OF COTTON. 65 The imports into the United States1 for each calendar year since 1885 have been as follows : Imports of Peruvian cotton into the United States. Year. 1885 1886 1887 18*8 1889 1890 Bales. Tear. Bales. 14 1891 10, 515 843 1892 13 000 2,493 1893 24 0U0 4,279 1894 19 000 7,650 1895 24, 000 9,500 AFRICA. Excluding Egypt, which has been treated in another place, the con- tinent of Africa produces a considerable amount of cotton on both the eastern and western coasts, as well as in the central part: in fact, the whole population is clothed in cotton, the greater part of which is home grown and manufactured. Cotton is indigenous to Senegambia, Liberia, the Congo States, and Soudan, and under proper cultivation these districts are capable of producing more cotton than is now raised in the United States, but such a condition of things is too remote a possibility to awaken any interest among cotton raisers or manufac- turers at the present time, and only mentioned to indicate a section where cotton might be abundantly and cheaply produced. EAST INDIES. Cotton is cultivated in all parts of Java, the estimated amount being about G,500 bales of 400 pounds each, one-half of which is used in Java for stuffing mattresses, cushions, etc. Some cotton is also grown in the Siam Malay States and is manufactured by the very primitive homemade machines into a cloth very durable, but not as attractive as the cotton cloths of Europe, which are largely superseding the domestic product among the natives. There has been some exportation of cotton from the East Indies to the continent of Europe, but the amount is comparatively insignificant. WEST INDIES. The West Indies produced cotton long before it was cultivated in the cotton belt of the United States. For a while the total American sup- ply to Europe came from these islands. In 1801 25,000 bales were exported; in 1836, 20,000 bales; and since that date the crop exported has steadily decreased until scarcely a thousand bales are credited to these islands. This decrease is largely due to a greater profit coming from the cultivation of other crops, and there is no prospect as long as the present conditions exist for an increase of the cotton crop in the West Indies. 1903— No. 33- 1 Cotton Facts, December, 1895. 66 THE COTTON PLANT. THE LEVANT. Under this name the cotton raised in Greece and Turkey and her provinces is known in the European market. The total amount is small, probably not more than 20,000 bales of 400 pounds each, 25 per cent of which is used in the countries where produced, the remaining 75 per cent being shipped to England and the continent of Europe. It is safe to say that the acreage under cotton culture in this district is decreasing annually. Some of the staple is long, white, and equal to good American cotton. SOUTH SEA ISLANDS. Many of the islands of the South Sea or Southern Pacific produce some cotton of a good staple, but their output, considered as individual islands or as a combined section, is inconsiderable, having no apprecia- ble effect upon the supply in the markets of the world. Australia has also produced some cotton, and some sections of the country are very well adapted to the growth of the plant. Attempts have been made to extend its culture, but without success, owing prob- ably to the sparse population and the comparatively small returns from the cultivation of the fiber. BOTANY OF COTTON. By Walter H. Evans, Ph. D. Office of Experiment Stations. The cotton plant belongs to the Malvaceae, or the mallow family, and is known scientifically by the generic name Gossypinm. It is indigenous principally to the islands and maritime regions of the tropics, but under cultivation its range has been extended to 40° or more on either side of the equator, or to the isothermal line of 60° F. In this country lati- tude 37° north about represents the limit of economic growth. The generic description as compiled from various sources is as follows: Gossypium. — Herbaceous, shrubby or arborescent, perennial, but in cultivation herbaceous and annual or biennial, often hairy, with long, simple, or slightly branched hairs, or soft and tomentose, or hirsute, or all the pubescence short and stellate, rarely smooth throughout; stem, branches, x>etioles, peduncles,11 leaves, involucre, corolla, ovary, style, capsule, and sometimes the cotyledons more or less covered with small black spots or glands. Roots Ibaprooted, branching, long, and penetrating the soil deeply. Stem erect, terete, with dark-colored, ash- red, or red bark and white wood, branching or spreading widely. Branches terete or somewhat angled, erect or spreading, or in cultiva- tion sometimes very short. Leaves alternate, petioled, cordate or sub- cordate, 3 to 7 or rarely 9 lobed, occasionally some of the lower and upper ones entire, 3 to 7 veined. Veins branching and netted ; the mid vein and sometimes adjacent ones bear a gland one-third or less the distance from their bases, or glands may be wholly absent. Stipules in pairs, linear- lanceolate, acuminate, often caducous. Flowers pedunculate. Peduncles subangular or angular, often thickened toward the ends, short or very short, erect or spreading ; in fruit sometimes pendulous, sometimes gland- ular, bearing a leafy involucre. Involucre 3-leaved, or in cultivation sometimes 4; bracteoles often large, cordate, erect, appressed or spread- ing at summit, sometimes coalescent at base or adnate to the calyx, den- tate or laciniate, sometimes entire or nearly so, rarely linear. Calyx short, cup-shaped, truncate, shortly 5 dentate, or more or less 5-parted. Corolla hypogynous. Petals 5, often coalescent at base and by their claws adnate to the lower part of stamen tube, obovate, more or less unequally transversely dilated at summit, convolute in bud. Stamina! column dilated at base, arched, surrounding the ovary, naked below, above narrowed and bearing the anthers. Filaments numerous, filiform, simple or branched, conspicuous, exserted. Anthers kidney-shaped, 1-celled, dehiscent by a semicircular opening into two valves. Ovary 67 68 THE COTTON PLANT. sessile, simple, 3 to 5 celled. Ovules few or many, in two series. Style clavate, 3 to 5 parted ; divisions, sometimes erect, sometimes twisted and adhering together, channeled, bearing the stigmas. Capsule more or less thickened, leathery, oval, ovate-acuminate, or subglobose, mucro- nate, loculicidally dehiscent by 3 to 5 valves. Seed numerous, subglo- bose, ovate or subovate, oblong or angular, densely covered with cotton or rarely glabrous. Fiber sometimes of two kinds, one short and closely adherent to the seed; the other longer, more or less silky, of single simple flattened cells more or less spirally twisted, more readily separa- ble from the seed. Albumen thin, membranous, or none. Cotyledons Xflicate, auriculate at base, enveloping the straight radicle. Embraced by this description as synonyms are Sturtia li. Brown and Xylon Tournefort, both of which are antedated by the name Gossypium of Linnaeus. On account of their great variability the species of this genus are difficult of limitation, and various attempts have been made to classify them. Linnseus described at least 3 species, and since that time the number of species and synonyms has increased enormously. Two monographs of the genus have been published by Italian botanists, the first by Filippo Parlatore1 in 1866, in which the author recognized 7 species, with 8 others in doubt. The other monograph was by Agostino Todaro,2 published in 1877, in which are described 52 species, with 2 as uncertain. Hamilton sought to avoid confusion by dividing the genus into 3 species, the white seeded, black seeded, and yellow linted, to which he gave the names album, nigrum, and croceum. A recent publi- cation, Index Kewensis,3 recognizes 42 species, of which but a very few are of economic importance, and mentions 88 others that have been reduced to synonyms, most of them being synonyms of species in com- mon cultivation. The great variability and the tendency to hybridize make it difficult to determine to which species a given plant may belong. No cultivated plant responds so quickly to ameliorated con- ditions of soil, climate, and cultivation as the cotton plant, and to this fact is due much of the confusion as to species and varieties. Another factor entering into the confusion is the imperfectly known types that have been described as species. It has been stated that some of the species now widely cultivated are wholly unknown in a wild state, and some of the specimens described by Linnaeus were in all probabil- ity from plants that had long been in cultivation. The work of estab- lishing the origin of the cultivated species has been still further complicated by the exchange of seed from country to country th.it has been going on for at least four centuries. Among the species recognized to be of more or less economic impor- tance are G. arboreum, G. neglectum, G. brasiliense, G. herbaceum, G. 1 Le specie dei cotoni. 2Rel. snlla coltura dei cotoni in Italia, 1877-78. 3 Vol. 2, pp. 1057, 1058. BOTANY OF COTTON. 69 barbadense, and perhaps a few others. In this country only the her- baceous cottons are cultivated to any extent. The shrubby and arbo- reous are grown occasionally as curiosities, but they seldom or never produce any lint in regions having as low a mean temperature as the cotton belt of the United States. The determination of the species of cotton grown in this country pre- sents some peculiar difficulties. The authorities differ widely regarding the specific origin of the short-staple or upland cotton, while more Fig. 1. — Sea Island cotton (from photograph furnished by Mississippi Experiment Station). nearly agreeing on that of the Sea Island cotton. The latter is gener- ally considered as having originated from 0. barbadense, a technical description of which is given below. G. barbadense Linn, was originally described as having leaves 3-lobed, entire. A more amplified compiled description is as follows: Shrubby, perennial, 6 to 8 feet high, but in cultivation herbaceous and annual or biennial, 3 to 4 feet high, glabrous, dotted with more or less prominent black glands. Stem erect, terete, branching. Branches graceful, 70 THE COTTON PLANT. spreading, subpyramidal, somewhat angular, ascending, at length recurving. Leaves alternate, petiolate, as long as the petioles, rotund, ovate, subcordate, 3 to 5 lobed, sometimes with some of the lower and upper leaves entire, cordate, ovate, acuminate; lobes ovate, ovate- lanceolate, acute or acuminate, channeled above, sinus subrotund, above green, lighter on the veins, glabrous, beneath pale green and glabrous, 3 to 5 veined, the niidvein and sometimes one or both pairs of lateral veins bearing a dark-green gland near their bases. Stip- ules erect or spreading, curved, lanceolate-acuminate, entire, or some- what laciniate. Peduncles equal to or shorter than the petiole, erect, elongating after flowering, rather thick, angled, sometimes bearing a large oval gland below the involucre. Involucre 3-parted, erect, segments spreading at top, many-veined, broadly cordate- ovate, exceeding half the length of the corolla, 9 to 11 divided at top, divisions lanceolate, acuminate. Calyx much shorter than the involucre, bracts clip shaped, slightly 5-toothed or entire. Corolla longer than the bracts. Petals open, but not widely expanding after flowering, broadly obovate, obtuse, crenate or undulate margined, yel- low or sulphur colored, with a purple spot on the claw, all becoming purplish in age.* Stamens about half the length of the corolla, the tube naked below, anther bearing above. Style equaling or exceeding the stamens, 3 to 5 parted. Ovary ovate, acute, glandular, 3, rarely 4 to 5 celled. Capsule a little longer than the persistent involucre, oval, acuminate, green, shining, 3, rarely 4 to 5 valved. Valves oblong or ovate-oblong, acuminate, the points widely spreading. Seed G to 9 in each cell, obovate, narrowed at base, black. Fiber white, 3 to 4 or more times the length of the seed, silky, easily separable from the seed. Cotyledons yellowish, glandular punctate, Species which have been considered synonyms of G. barbadense and to which the above description will apply are G. frutescens Lasteyr., G. fuscum Roxb., G. glabrum Lam., G. jamaicense Macfad., G.javani- cum Blume, G. maritimum Todaro, G. nigrum Hamilton, G. oligosper- mum Macfad., G. perenne Blanco, G. perurianum Cav., G. punctatum Schum. andThonn., G. racemosum Poir., G.religiosum Parlatore, Q.riti- folium Roxb., and perhaps others. This species is indigenous to the Lesser Antilles and probably to San Salvador, the Bahamas, Barbados, Guadaloupe, and other islands between 12° and 26° north latitude. By cultivation it has been extended throughout the West Indies, the maritime coast of the South- ern States, Central America, Puerto Rico, Jamaica, etc., southern Spain, Algeria„the islands and coast of western tropical Africa, Egypt, Island of Bourbon, East Indies, Queensland, New South Wales, etc. It may be cultivated in any region adapted to the olive and near the sea, the principal requisite being a hot and humid atmosphere, but the results of acclimatization indicate that the humid atmosphere is not entirely nec- essary if irrigation be employed, as this species is undoubtedly grown BOTANY OF COTTON. 71 extensively in Egypt. As a rule, the quality of the staple increases with the proximity to the sea, but there are exceptions to this rule, as that grown on Jamaica and some other islands is of rather low grade, while the best fiber is produced along the shores of Georgia and Carolina. According to Royle1 "the quality is influenced not only by tempera- ture, but the balance between the amount of moisture taken up by the roots and that given off by the leaves must be considered, as well as the varied processes of culture and choice of varieties suited to each particular locality." This observation applies to all kinds of cotton, and not to the Sea Island alone. Some authors question the American origin of this species, and one, Maxwell T. Masters,2 claims that it is of central African origin ; but the weight of testimony is against him, and in all probability this was the species grown on the island of San Salvador at the time of the landing of Columbus and by him carried to Spain. Some authors claim to recognize a difference between G. barbadense and 67. maritimum; but whether it is more than a cultural variation is very difficult to determine, and for our purpose they are considered as botanically synonymous, while they may be commercially different. There is a well-marked form of the Sea Island cotton.to which Todaro gave the varietal name of polycarpun to which is usually referred the Bamia variety of Egyptian cotton. It is principally characterized by numerous flowers springing from a single axil, and an erect, slightly branching habit, hence giving a large yield per acre. On poor soil it soon degenerates to an ordinary form of Sea Island. This is considered by Sir J. D. Hooker3 as a well-marked seminal sport, with a fastigiate habit, from some kind of Egyptian cotton, the bulk of which belongs to the Sea Island form of G. barbadense. In one of the Kew reports1 the idea that Bamia is a hybrid between okra and cotton is shown to be incorrect. The cultivation of Bamia in Egypt is said to require more irrigation than the ordinary kinds. The yield of lint from Sea Island cotton is less than that from any other kind grown in this country, but on account of the length and quality of the fiber it is adapted to uses to which the other kinds are not suited, and its high market value compensates for the small yield. The botanical species from which have been developed the multitudi- nous forms of upland cotton is of less certain restriction. Scarcely any of the authors agree in some of the most important particulars when discussing its origin. However, the weight of opinion seems to be that the species is either G. herbaceum or G. hirsutum, and as these are con- sidered synonyms their description is combined under the name of the former. 1 Cultivation of Cottoii in India. 2 Jour. Linn. Soc, XIX, p. 213. 3 Flora of British India. 4 Kew Report, 1887, p. 26. 72 THE COTTON PLANT. Gr. herbaceum Linn. — Shrubby, perennial, but in cultivation herba- ceous, annual or biennial. Pubescence variable, part being long, simple or stellate, horizontal or spreading, sometimes short, stellate, abundant, or the plants may be hirsute, silky, or all pubescence may be more or less wanting, the plants being glabrous or nearly so. Glands niore or less prominent. Stem terete, or somewhat angled above, branching. Branches spreading or erect. Leaves alternate, petioled, the petioles Fie. 2.— Upland cotton (from, photograph furnished jy Georgia Experiment Station). about equaling the blades, cordate or subcordate, 3 to 5, rarely 7-lobed. Lobes from oval to ovate, acuminate, pale green above, lighter beneath, more or less hairy on the veins, 3 to 5 or 7-veiued, the midveiu and some- times the nearest lateral veins glandular toward the base or glands wanting. Sinus obtuse. Lower leaves sometimes cordate, acuminate, entire, or slightly lobed. Stipules erect or spreading, ovate-lanceolate to h'near lanceolate, acuminate, entire oi occasionally somewhat dentate. Peduncles erect in flower, becoming pendulous in fruit. Involucre 3, BOTANY OF COTTON. 73 rarely 4 parted, shorter than the corolla, appressed spreading in fruit, broadly cordate, incisely serrate, the divisions lanceolate, acuminate, entire or sometimes sparingly dentate. Calyx less than half the length of the involucre, cup-shaped, dentate, with short teeth. Petals erect, spreading, obovate or cuneate, obtuse or emarginate, curled or crenu- late, white or pale yellow, usually with a purple spot near the base, in age becoming reddish. Stamens half the length of the corolla. Pistil equal or longer than the stamens. Ovary rounded, obtuse or acute, glandular, 3 to 5 celled. Style about twice the length of the ovary, 3 to 5 parted above, the glandular portion often marked with 2 rows of glands. Capsule erect, globose or ovate, obtuse or acuminate, inucro- nate, pale green, 3 to 5 celled. Valves ovate to oblong, with spreading tips. Seed 5 to 11 in each cell, free, obovate to subglabrous, narrowed at base, clothed with two forms of fiber, one short and dense, closely enveloping the seed, the other 2 to 3 times the len gth of the seed white, silky, and separating with some difficulty. Cotyledons somewhat glandular punctate. This species includes in its synonyms the following : G. album Hamil- ton, G. chinense Fisch. and Otto, G. croceum Hamilton, G. eglandulosum Cav., G. elatum Salisb., G. glandulosum Steud, G. hirsutum Linn., G. indicum Lam., G. latifolium Murr., G. leoninum Medic, G. macedonicum Murr., G. micranthum Cav., G. molle Mauri, G. nanking Meyen, G. obtusifolium Roxb., G. paniculatum Blanco, G. punctatum Guil. and Perr., G. religiosum Linn., G. siamense Tenore, G. sinense Fisch., G. strictum Medic, G. tricuspidatum Lam., and G. vitifolium Roxb., together with numerous others the descriptions of which are too indefinite or the specimens too meager to determine them positively. The origin of this series is much more confused than that of the Sea Island cotton. If we separated the upland cotton into two species, viz, G. herbaceum and G. hirsutum, probably the question would no doubt be simplified, as the former is generally considered of Asiatic origin, while the other is attributed to America. Todaro 1 claims that the form called by him G. hirsutum originated in Mexico, from whence it has been spread by cultivators throughout the warmer portions of the world. To this form he ascribes the Georgia upland cotton or the long- staple upland cotton. Parlatore2 considers it indigenous to some of the islands of the Gulf of Mexico as well as the mainland, and all green- seeded cotton, which is cultivated so widely, as originating from this form. On the other hand, he claims3 India, especially the shores of Coromandel, as the primitive home of G. herbaceum, from which place it has spread as extensively as its western congener, and is found in cultivation in nearly the same regions. Todaro says4 that G. herba- 'Eel. Bulla coltura dei cotoni in Italia, 1877-78, p. 212. 2Le specie dei cotoni, p. 43. ■! Ibid., p. 33. 4 Eel. sulla coltura dei cotoni in Italia, p. 132. 74 THE COTTON PLANT. ceum is spontaneous in Asia and perhaps also in Egypt, and he claims G. wightianum as the primitive form of the Indian cottons. Maxwell T. Masters ' claims G. stoclcsii as the original of all the cultivated forms grouped under G. herbaceum. Others consider G. herbaceum as a native of Africa, and it seems impossible from the mass of conflicting evidence to determine just where it did originate. It seems probable that G. her- baceum is not a definite species, but one developed by cultivation from, perhaps, several wild species, and it represents not a species but a group of hybrids and forms more or less closely related. However, if we consider it as a definite species, no violence is done in claiming it as indigenous to as widely separated regions as America and Asia, nor is it necessary to assume the migration from one place to the other, but rather to consider it an example of simultaneous evolution of a species in opposite portions of the globe without any communication between the progenitors of the species, examples of which are well known. It is a case of evolution influenced by like conditions, and it is probable that the needs of man entered very largely into the development by the selection of certain characters which it was desirable to perpetuate. The cottons usually called "nankeen" are only color variations of the above, and may be found in nearly every species that is cultivated. Authorities agree that in all probability the yellow lint is the wild form of all cottons, and this character can not be used to designate species. Species of secondary importance that may be occasionally met in this country, or species thought to be worthy of introduction, although their value is certainly less than those furnishing the staple crop, are as follows: G. arboreum Linn. — Shrubby perennial, but in cultivation sometimes annual or biennial; tomentose with two forms of hairs, one long and simple, the other more numerous, shorter, and stellate; glands small, scarcely prominent, more or less scattered. Stem erect, terete, very branching. Branches spreading, terete. Leaves alternate, petiolate, with petioles a little shorter than the blade, subcordate, 5 to 7 lobed, lobes oblong-lanceolate or lanceolate-acuminate, bristle-tipped, scarcely channeled above; sinus obtuse, often with a small lobe in some of the sinuses, beneath pale green and softly pubescent, 5 to 7 veined, the mid- vein and often the two adjacent ones with a reddish-yellow gland near their base; upper leaves palmately 3 to 5 lobed, lobes short. Stipules erect, spreading, lanceolate, acuminate. Peduncles axillary, erect before and spreading or horizontal after flowering and drooping in fruit, about three-fourths the length of the petioles, terete, destitute of glands, 1 to 2, usually 1-flowered, jointed above the middle, bearing a small leaf and 2 stipules at this point. Involucre 3-parted, appressed or scarcely spreading at summit, many-nerved, broadly and deeply cordate, ovate, acunr.nate, 5 to 9, rarely 3 dentate or nearly entire. Calyx much shorter 1 In Hooker's Flora of British India. BOTANY OF COTTON. 75 than the bracts, subglobose, truncate, crenulate or subdentate, with, a large gland at the base within the involucre. Corolla caiupanulate, petals erect or spreading, broadly cuueate, subtruncate, crisp or crenu- late, purple or rose colored, with a large, dark-purple spot at the base. Stamina! tube about half the length of the corolla. Pistils equaling or a little longer than the stamens. Ovary ovate, acute, glandular, usually 3 celled. Style little longer than the ovary, 3-parted, without glands. Capsule pendulous, a little longer than the persistent involucre, ovate rounded, glandular, 3 to 4 celled, and valved. Valves ovate-oval, spreading, mucronate-acuininate, the mucro recurved. Seed ~> to G, ovate, obscurely angled, black. Fiber two forms, one white, long, over- lying a dark-green or black down; not readily separable from the seed. This species of cotton appears to be indigenous to India and the regions bordering on the Indian Ocean. According to Watt,1 it is found near temples and in gardens, where it is said to be in flower most of the year. The plant is a perennial, lasting for five or six years or longer, and is not used as a field crop. The liber is fine, silky, and an inch or more in length, but little of it is produced. The cultural name given it is Nurma or Deo cotton, and its use is said to be restricted to making thread for the turbans of the priestly class. Its value is said to be greatly overrated. This species is sometimes known as G. religiosum. G. neglectum Tod. — Stem ereet. Branches slender, graceful, spread- ing. Leaves, lower ones 3 to 7 palmately lobed, segments lanceolate, acute, rarely bristle-tipped, sinus rounded, the small lobes in the sinuses less distinct than in the previous species, upper leaves 3-parted. Stip- ules next the peduncles semiovate, dentate, the others linear-lanceo- late, acute. Peduncles with short lateral branches, 2 to 4 flowered. Involucral bracts coalescent at base, deeply and acutely laciniate. Petals less than twice the length of the involucral bracts, obovate. unequally cuneate, yellow, with a deep-purple spot at base. Stamen- tube half the length of the corolla, naked at base. Capsule small, ovate, acute, cells 5 to 8 seeded, seed obovate, small, clothed with two forms of fiber, one very short, closely adherent, and of an ashy -green color, the other longer, rather harsh, white. This species, indigenous to India, is very similar to G. arboreum, and by some is thought to be a hybrid between that species and some other, or it may be only a cultural form of the first. It is a large bush, although sometimes only 18 inches in height, aud is extensively grown in India as a field crop. It is the Dacca cotton of Royle and Roxburgh and the. China cotton of the same authors. This species is cultivated in Bengal, the Punjab, and the Northwest Provinces, and it constitutes to a large extent the Bengal cotton of commerce.2 Todaro^ has sepa- rated from the species two varieties — roxburghianum and chinense — 1 Diet. Economic Products of India, Vol. IV, p. 5. 2 Ibid., p. 7. 3 Eel. sulla coltura dei cotoni in Italia, p. 169. 76 THE COTTON PLANT. corresponding' to the Dacca and China cottons above mentioned. It is very probable that both the varieties and the species are not well founded, but are cultural forms. There is another Indian species, G. wiglitianum Tod., that is claimed to be the form chiefly cultivated in India. It greatly resembles the G. herbaceum of India, but differs from that species in that the latter has broader and more rounded leaves, and broader, thinner, and deeper cut bracteoles. It is characterized as follows : G. ivif/htianum Tod. — Stems erect, somewhat hairy, branches spread- ing and ascending. Leaves when young densely covered with short, thick, stellate hairs, becoming nearly glabrate in age; ovate-rotund, scarcely cordate, 3 to 5 rarely 7 lobed, lobes ovate, oblong, acute, con- stricted at base into a rounded sinus. Stipules on the peduncles almost ovate, others linear-lanceolate, acuminate. Flowers yellow with a deep-purple spot at base, becoming reddish on the outside in age. Bracteoles small, slightly united at base, ovate, cordate, acute, shortly toothed. Peduncles erect in flower, recurved in fruit, one-fourth the length of the petioles. Capsule small, ovate, acute, 4-celled, with 8 seeds in each cell. Seeds small, ovate, subrotund, clothed with two forms of fiber, the inner short and closely adhering, other longer, white or reddish. This species is said to readily hybridize with G. neglectum, and numer- ous species have been founded upon these cultural forms. Among these hybrids are some of the most valuable of Indian cottons. The typical forms of the foregoing species of cotton have their seed free from each other, but there is another group in which the seed of each cell are closely adherent in an oval mass, from which appearance they are called " kidney " cottons. Most if not all these species are trop- ical, and their presence in this country as anything more than curiosities is highly improbable. The most important of them is G. braziliense Macfad., and in addition to the fact of the seed adhering in clusters the species is an arborescent plant with very large, 5 to 7 divaricate lobed leaves and very deeply laciniate involucral bracts. The cottons of South America, known to the trade as Pernambuco, Ceara, Santos, etc., are evidently not of this species, but belong to the G. barbadense and G. herbaceum series. The physiology and histology of the cotton plant seem to be subjects that have been almost wholly ignored by investigators, and the only references to these subjects are for the most part very general. According to Heuze,1 the time required for the maturity of a cotton crop is divided as follows: From seeding to flowering, New Orleans 80 to 90 days, Sea Island 100 to 110 days; from flowering to maturity, New Orleans 70 to 80 days, and Sea Island about 80 days, making the total period of growth about 5 to 6£ months. According to the same authority, the best average daily temperature for the growth of cotton 1 Plantes Industrielles, Vol. I, p. 139. BOTANY OF COTTON. 77 is from 60° to 68° F. for the period from germination to flowering and from 08° to 78° from flowering to maturity. Dr. Wight * says that for the proper maturity of the best qualities of American cotton an increas- ing temperature during the period of greatest growth is required. The failure to produce in India a quality of fiber equal to the American product from the same kind of seed is attributed to the fact that in the climate of the former there exists a diminishing rather than an increas- ing average daily temperature. The effect of too much rain is to form too much plant and not enough fruit, while serious drought causes a stunted growth of the plant in which few bolls are formed and these ripen prematurely. In the latter case the resultant crop is generally short in staple and poor in quality. The structure of the cotton fiber has been studied to a considerable extent, and the works of Bowman 2 and Monie 3 may be considered as standard on this part of the subject. The first thing noticed in com- paring samples of cotton is the difference in the length and the fineness of the fiber, and upon these factors almost entirely depends the com- mercial grading of the crop. The principal species of cotton vary in respect to the length of their fiber within rather constant limits, dependent upon soil, culture, and atmospheric conditions. The follow- ing table compiled from numerous measurements taken during a period of years shows the maximum, minimum, and average length of fiber for some of the more important varieties, and also the average diameter of the same: Length and diameter of the principal cotton fibers. Variety. Sea Island New Orleans Texas Upland Egyptian Brazilian Indian varieties : Native American seed. Sea Island seed Length of staple. Average — diameter Maximum. Minimum. Average. | of staple. Inches. 1.80 1.16 1.12 1.06 1.52 1.31 1.02 1.21 1.65 Inches. 1.41 .81 1.30 1.03 .97 .95 1.36 Inches. 1.61 1.02 1.00 .93 1.41 1.17 1.08 1.50 Inch. . 000640 .000775 . 000763 .000763 .000655 .€00790 .000844 . 000825 . O0U73U From the above table it will be seen that, as a rule, the longer the fiber the less its diameter. The extreme variation in length of the above fibers, from the figures as shown in the table, is from 0.25 to 0.30 inch. In proportion to their size the variation in diameter is much greater than that shown for the length. If a very immature boll be cut transversely the cut section will show that it is divided by longitudinal walls into three or more divisions, •Jour. Agr. Hort. Soc. India, 7 (1849-50), p. 23. structure of the Cotton Fibre, F. H. Bowman, Manchester, 1881. 3 The Cotton Fibre, its Structure, etc., Hugh Monie, Manchester and London, 1890. 78 THE COTTON PLANT. and the seed will be shown attached to the inner angle of each division. The seed retain this attachment until they have nearly reached their mature size and the growth of lint has begun on them, when their attachments begin to be absorbed and by the increased growth of the lint tlie seed are forced to the center of the cavity. The development of the liber commences at the end of the seed farthest from its attach- ment, and gradually spreads over the seed as the process of growth con- tinues. The first appearance of the cotton fiber occurs a considerable time before the seed has attained its full growth, and commences by the development of cells from the surface of the seed. These cells seem to have their origin in the second layer of cellular tissue, and force them- selves through the epidermal layer, which seems to be gradually absorbed. The cells which originate the fiber are characterized by the thickness of their cell walls when compared with their diameter. The method of growth, according to Bowman,1 is by the successive linear development of cells, the walls of which are absorbed at the point of contact until an elongated cell is produced, which constitutes the cot- ton fiber. The continued growth of this mass of fiber assists in burst- ing open the pod when the period of maturity is reached. The length of the fiber varies considerably on different parts of the seed, being longest on the crown and shortest at the base. It is claimed that the fibers do not attain their full length until the pod has been ox>ened and the fibers are exposed to the drying and ripening effect of the air and sun. In their earliest stages the young fibers appear circular in section, but with their increase in length the walls become thinner and finally collapse into a flat, thin-walled fiber, iu appearance like a thin, trans- parent ribbon. With the opening of the boll there is a rapid consoli- dation of the liquid cell contents, which by being deposited on the inner side of the walls give to the fiber a greater thickness and density. As the degree of maturity is increased the fiber once more becomes rounded in section. As this action is not perfectly regular, owing to the unequal pressure and deposition of the cell contents, the fibers become twisted, a character readily recognized under the microscope, and one that distinguishes cotton from any other fiber. In the early period of their formation the cells are filled with astrin- gent juices whose presence may be recognized by applying the tongue to the cut surface of an immature boll. During the process of ripening these juices are replaced by others of a neutral or saccharine nature, and when perfectly ripe th ~ cotton fiber consists almost entirely of cel- lulose. When viewed under a microscope the general appearance of a cotton fiber is that of an irregular, flattened, and somewhat twisted tube, the tubular form sometimes being lost in the completely flattened fiber. The edges of the fiber are somewhat thickened and slightly corrugated. 1 Structure of the Cotton Fibre, p. 25. BOTANY OF COTTON. 79 The hollow tubular character aud constant diameter of the fiber are maintained for about three-fourths its length, when it tapers to a point, where it is perfectly cylindrical and often solid. From various causes there are often found solid places in the body of the fiber, and where such places exist the quality of the staple is reduced, owing to the ine- quality with which such fibers take up dyestuffs. The twist in the fiber, which seems to be an acquired character not possessed by wild cotton, is explained by Monie ' as follows: The rotary motion begins with the process of vacuation in the fiber, caused by the withdrawal of some of the fluid in the fiber when the seed begins to ripen, and as this is effected slowly and progressively, beginning near the extremity farthest from the seed and gradually receding toward the base, the free end or point becomes twisted on its own axis several times, thus producing the convoluted form exhibited under the microscope. In every lot of cotton three classes of fibers may be recognized — (1) unripe, (2) half-ripe, and (3) ripe. These conditions are dependent upon several factors, the most important of which is the gathering of cotton before it has been exposed for a sufficient time to the ripening action of the air aud sun. The other cause is due to the different stages of maturity of the filaineuts on different parts of the same seed. Unripe cotton when examined with the aid of a microscope appears extremely thin and transparent, and usually with little or no twist, and it is of little use for manufacture. When used it contracts and curls up in the warm atmosphere of the factory, causing yarn spun from cotton con- taining much unripe fiber to depreciate greatly in value. The half-ripe fiber has the same characters, but to a lesser degree, and is more valu- able than the former, but it is only the ripe cotton fiber that possesses all the requisites for perfect spinning and dyeing. " The differences of the three kinds of fibers as observed under the microscope are shown in the accompanying figures redrawn from Bowman's Structure of the Cotton Fibre. A perfect cotton fiber consists of four parts — (1) an outer membrane, which constitutes the outside skin of the fiber; (2) the real cellulose, which constitutes 85 per cent of the fiber; (3) a central spiral deposit of a harder nature than the rest of the fiber, and (4) a central secretion that corresponds somewhat to the pith of a quill. Covering the cotton fiber is a sort of varnish or oleaginous deposit, technically known as cotton wax. This is said by Monie2 to amount to about 2 per cent by weight of the fiber, and must be removed before the yarn is dyed, otherwise the coloring will be poorly done. The presence of this substance on the fiber is readily shown by the difficulty with which ordinary cotton absorbs moisture. Absorbent cotton is cotton that has been treated in such a way that all the cotton wax is removed. The details of treatment that the fiber must be subjected to before 'The Cotton Fibre, its Structure, etc., p. 25. 2 Ibid., p. 24. 80 THE COTTON PLANT. spinning and dyeing belong to the technique of those trades, and would be out of place in this article. The theory which explains the power of Fig. 3. — Cotton fibers in longitudinal and cro^s section: AAA, unripe fibers; BB, half-ripe fibers; CCC, fully ripe fibers. fibers to take up and retain dyes seems not very well authenticated, and is intentionally omitted. CHEMISTRY OF COTTON. By J. B. McBkyde, Chemist of the Tennessee Experiment Station, and W. H. Beal, Office of Experiment Stations. As a rule our staple agricultural plants have not received the thor- ough, systematic chemical investigation that their importance demands. It is true that in many cases the commercial products have been the subject of numerous and sometimes exhaustive chemical studies, but for the plant as a whole, especially with reference to its composition and demands upon the soil at different stages of growth, the analytical data are singularly incomplete and unsatisfactory. This is strikingly true of the cotton plant, of which it has been said: "It is more than probable that less is known of the composition of this plant, with the exception of the seed, than of any other of our staple crops." Although a number of recent studies of the cotton plant have been reported which in a measure supply this deficiency, it will be found that of some 1,500 analyses of the plant and its various parts reported in this article by far the greater number relate to the seed and its products, the rest of the plant, if we except the lint, having usually been ignored. In the present article chemical data relating to cotton have been col- lected from every available source and classified and arranged for the use of investigators, little technical discussion of the data being attempted except where it is necessary to emphasize a point not clearly brought out by the tables. As a rule the data used in the text tables have been limited to maxima, minima, and averages of the more impor- tant constituents, individual analyses being given only when necessary to illustrate some particular point. A complete compilation of analy- ses will be found in tables at the end of the article. The order of arrangement is (1) the entire plant and (2) its different parts, beginning with the roots. Under each of the main heads the order is (1) fertiliz- ing constituents, (2) proximate constituents, and (3) miscellaneous chem- ical studies. This arrangement allows discussion in logical sequence, first, of the chemical data relating to the demands upon the soil and the growth and development of the plant; and second, that relating to the character of the product and its utilization. 1993— No. 33 6 81 82 THE COTTON PLANT. ENTIRE PLANT. FERTILIZING CONSTITUENTS. The results of available analyses of the entire plant, including the roots, are given in tbe following table: Fertilizing constituents of cotton {entire plant) . No. Tate of analy- sis. Water. Ash. Nitro- gen. Phos- phoric acid. Potash. Lime. Mag- nesia. 1 2 Tonng plants collected June 3 (with two leaves) a Young plants collected June 23 a 1890 1890 1857 1890 Per ct. 7.36 Per ct. 17.36 16.21 4.57 5.81 Per ct. 3.82 3.93 1.46 Per ct. 3.51 2.10 .44 .44 Per ct. 2.69 1.90 M. 17 1.32 Per ct. 5.34 4.70 .82 1.42 Per ct. 1.29 1 15 3 .46 4 do .52 a Water free. b Potash and soda. The data are too limited and unsatisfactory to warrant averages. The analyses of young plants collected June 23 show a decrease in the percentages of ash constituents, especially phosphoric acid, from those found in plants collected twenty days earlier. In connection with investigations on the effect of different fertilizers on the composition of cotton grown on poor and fertile soils, Anderson1 has made analyses of the above ground portion of cotton plants in the flowering and boiling stages, the principal results of which are given in the table below : Fertilizing constituents in dry matter of cotton plants grown with different fertilizers on poor and fertile soil. Potash. Phosphoric acid. Nitrogen. Flower- ing stage. Boiling stage, a Flower- { Boiling ing stage, stage. a Flower- ing stage. Boiling stage.a Poor soil: Per cent. 2.03 Per cent. 1.26 Per cent. 0.93 .86 Per cent. 0.79 Per cent. 3.49 3.91 3.38 3.84 3.86 3.69 3.97 3.65 3.99 3.98 3.72 Per cent. 1 88 2.75 .78 .63 .70 .92 .83 .83 .83 .96 .91 .81 .85 .80 .86 2.14 1.82 2 2.55 3.14 3.29 3.32 3.23 2.98 3.20 3.10 3.61 2.12 1.05 2. 12 2.56 2.61 .35 .54 .49 .56 .74 1 97 Nitrogen and phosphoric acid 1.88 1 84 Nitrogen, potash, and phosphoric 1.83 fertile soil: 2.31 2.03" 1.49 2.75 3.05 3.90 .74 .69 .90 .70 3. 83 2 , 1.49 13 Collected Aug. 17 1.75 11 Collected Aug. 28 2.25 7.23 3.01 4.43 9.89 2.92 5.60 48.57 38.55 44.19 39.15 29.14 33.92 2.77 .96 2.04 a Apsumed. The crude fiber fluctuated both years, though the tendency was to increase. The protein fluctuated also, aud was very different for the two seasons. In 1890 it remained nearly the same throughout the season, while in 1891 the tendency was rapidly downward, being at the end about one-third what it was at the beginning and about half what it was at the close of 1890. The ether extract and carbohydrates also fluctuated, both being higher, as a rule, the latter season. The ash fluctuated a little both years, but the tendency was to become less. At the close it was half what it was at the beginning of the season.1 MEDICINAL PROPERTIES. Cotton-root bark (Gossypii radicis cortex TJ\S. P.) contains a chemical substance similar in its action to ergot. The active property appears to reside in a red resin, but as far as can be ascertained no separated principle, representing the full activity of the bark, has yet been ex- tracted from the drug. E. S. Wayne in an article entitled " Medicinal properties of cotton roots" 2 gives ihe results of a study of the chemical properties of this drug. STEMS. FERTILIZING CONSTITUENTS. The stems constitute about 23 per cent of the entire plant. The fol- lowing table shows the fertilizing constituents of stems at various stages of growth. The maxima, minima, and averages are compiled from 20 analyses. 1 Mississippi Sta. Tech. Bui. 1. 2Amer. Jour. Pharm., 1872, p. 287. CHEMISTRY OF COTTON. 87 Fertilizing constituents of cotton stems. No. Date of analy- sis. Water. Ash. Nitro- Phos- plioric acid. Potash. Lime. Mag- nesia. Collected Collected Collected Collected Collected Collected Collected Collected Collected Collected Collected Collected Collected Collected July 2.. July 12. July 22. Aug. 1.. Aug. 11. Aug. 22. June 18. J une 28. JulyS.. July IS. July 28. Aug. 7.. Aug. 17. Aug. 28. 1890 1890 1890 1890 1890 1890 1891 1891 1891 1891 1891 1891 1891 1891 Per ct. a 10 a 10 a 10 a 10 a 10 a 10 a 10 a 10 a 10 a 10 alO alO alO alO Per 6. 5. 6. 5. 3. 2. 9^ 6. 5. 5. 4. 4. 4. 3. Per ct. 1.66 1.66 1.59 1.60 1.03 .96 3.28 1.76 1.78 1.54 1.68 1.63 1.10 1 Per ct. 0.66 .71 .51 .47 .51 .26 1.08 .76 .78 .74 .88 .67 .64 .49 Per ct. 1.79 1.18 1.49 1.36 1 .68 3.91 2.35 1.93 1.76 1.34 1.25 1.43 1.04 Per ct. 1.48 1.28 1.30 1.12 .83 .56 1.32 1.13 .94 .88 .85 .82 .73 .83 Per ct. 0.68 .67 .63 .67 .44 .31 .72 .50 .47 .47 .35 .31 .22 C Minimum. 20 analyses < Maximum { Average . . 8.41 11.71 10.01 2.34 9.64 4.80 .74 3.28 1.46 .59 .26 3.91 1.41 .50 1.52 .97 .42 a Assumed. The same general tendencies observed in case of the roots are noticed here, but the variations are more irregular. PROXIMATE CONSTITUENTS. The table below gives 15 proximate analyses of the stems. Analyses from 1 to 14, inclusive, show the composition of the stem at different stages of growth, and No. 15 is of fully matured plants. Proximate constituents of cotton stems. No. Date of analy- sis. Water. Ash. Protein. Fiber. Nitrogen- free extract. Pat. 1 Collected July 2 1890 1890 1890 1890 1890 1890 1891 1891 1891 1891 1891 1891 1891 1891 1889 Per cent. a 10 a 10 alO a 10 a 10 a 10 alO a 10 alO a 10 a 10 a 10 a 10 a 10 10.06 Per cent Per cent. 6.63 10.40 Per cent. 40.37 42.81 41.65 48.17 47.36 49.44 32.51 37.56 34.64 33.32 39. 95 Per cent. 31.04 30.27 30.77 25.89 30.22 29.75 25.59 33.27 36.43 39.87 31.61 33.15 34.88 35.51 35.04 Percent. 1.56 fl Collected July 12 5.39 6.10 5.41 3.93 2.75 9.64 6.44 5.31 5.22 4.44 4.27 4.07 3.92 4.09 10.40 9.94 10.02 6.49 6.04 20.45 11.02 11.13 9.60 10.50 1.13 S Collected July 22 1.54 4 1.41) 5 Collected Aug. 11 2.00 6 Collected Aug. 22 2.02 7 1.81 8 Collected June 28 1.71 q Collected July 8 2.49 10 1 99 11 Collected July 28 3 50 T>, 10.16 : 36.61 6. 87 42. 54 2.21) 1.64 13 14 15 Collected Aug. 28 6.25 4.90 42.71 45.16 1.61 .81 9.64 2.75 5.23 20.45 4.90 9.54 49.44 32.51 40.77 39.87 25.59 32.63 3.50 .81 10 1.83 a Assumed. Though it fluctuated, the crude fiber in the stem increased as the plant matured. The protein fluctuated, but the tendency was to decrease, and this was very marked at the later periods of growth, when the bolls were forming. The ether extract fluctuated, increasiug at the close in 1890 and decreasing at the same period in 1891. The carbohydrates remained nearly constant the first year throughout the period of growth, but increased slightly during the latter period as the plant grew older. ' 1 Mississippi Sta. Tech. Bui. 1. 88 THE COTTON PLANT. LEAVES. FERTILIZING CONSTITUENTS. The leaves have been found to be a little over 20 per cent of the plant. The following table gives 16 analyses of the mineral matter of the leaves. JSTos. 1 and 2 are from mature plauts. Nos. 3 to 16 are from the same samples reported in previous tables, and show the chemical changes taking place iu leaves during a part of their period of growth : Fertilising constituents of cotton leaves. No. Leaves of mature plants Leaves of mature plants Collected July 2 Collected July 12 Collected July 22 Collected Aug. 1 Collected Aug. 11 Collected Aug. 22 Collected June 18 Collected June 28 Collected July 8 Collected July 18 Collected Jul v 28 Collected Aug. 7 Collected Aug. 17 Collected Aug. 28 i Minimum . 19 analyses < Maximum ( Average .. Date of analy- sis. 1890 1890 1890 1890 1890 1890 1890 1891 1891 1891 1891 1891 1891 1891 1891 Water. Per ct 9.50 12. 14 a 10 a 10 a 10 a 10 a 10 a 10 a 10 a 10 «10 a 10 . a 10 a 10 a 10 a 10 9.50 12.14 10.10 Ash. Per ct. 16. 42 12. 05 13. 68 14.09 14.99 15.32 12.33 13.08 11.92 12.17 11.38 11.57 13.73 11.86 11.93 10.73 9.33 17.26 13.11 Nitro- gen. Per ct. 2.37 2.45 3.80 3.92 3.80 3.65 3.21 3.11 4.27 3.87 3.73 3.55 3.53 !-;.35 3.03 2.67 1.41 4.27 3.21 .42 2.08 1.19 Pot- ash. Per ct. 0.83 1.33 2.63 1.58 1.48 1.27 1.21 1.65 2.36 2.54 1.93 2. 21 L99 1.59 1.89 1.72 .83 2.63 1.80 Lime. Per ct. 7.08 4.10 5.28 5.31 5.63 5.27 4.61 4.66 3.83 3.65 3.68 3.75 4.87 3.89 3.56 3.67 2.71 7.08 4.44 Mag- nesia. Per ct. 1.26 .76 1.26 1.23 .99 1.28 .98 1.03 .66 .66 .81 .77 .93 .68 .60 .82 .07 1.28 .87 a Assumed. That there is slight decrease of total ash as the period of growth advances is indicated, although this is by no means regular. There is also a slight though fluctuating tendency to a decrease in all of the individual fertilizing constituents except phosphoric acid, which shows irregular but considerable increase. PROXIMATE CONSTITUENTS. The table below gives 15 proximate analyses of the leaves at various stages of growth; they show the chemical changes which take place as the plant advances in maturity. CHEMISTRY OF COTTON. 89 Analysis No. 1 is of the leaves of mature plants, as may be seen by the results : Proximate constituents of cotton leaves. No. Date of analy- sis. Water. Ash. Protein. Fiber. Nitrogen - free extract. Per cent. 43.32 34.51 33.89 34.64 29.72 35.13 37.37 36.32 37.10 37.04 37.54 35.14 37.93 38.52 44.16 Fat. 1 2 3 CollectedJuly 2 Collected July 12 1889 1890 1890 1890 1890 1890 1890 1891 1891 1891 1891 1891 1891 1891 1891 Per cent. 10.82 a 10 a 10 a 10 a 10 alO a 10 a 10 a 10 a 10 a 10 a 10 alO a 10 a 10 Per cent. 14.24 13.68 14.09 14.99 15.32 12.33 13.08 11.92 12. 17 11. 39 11.57 13.73 11.86 11.93 10.73 Per cent. 15.06 23.76 24.53 23.77 22.84 20.08 19.46 26.64 24.19 23.29 22.17 22.06 20.94 18.98 16.74 Per cent. 10.04 13.69 13.19 11.74 16.73 18.15 13. 04 11.15 11.45 11.39 12.51 11.29 12.79 11.93 9.50 Perct. 6.52 4.36 4 30 4 Collected J uly 22 4 86 5 6 Collected Aug. 11 4.31 7 Collected Aug. 22 8 g Collected June 28 5 09 10 11 Collected July 8 Collected July 18 6.89 6 21 12 13 CollectedJuly 28 7.78 6 48 U Collected Au". 17 8 64 Li Collected Aug. 28 8.87 15.32 i 26.64 10. 73 i 15. 06 12. 87 21 . 64 18.15 9.50 12.57 44.16 29.72 36.82 8 87 3 97 6 05 a Assumed. The proportionof crude liber in the leaves was small throughout the entire periods of growth, though it fluctuated both seasons; it fell off rapidly at the last period botb years. The protein was very high in the leaves, but continually became less as the plant matured. The ether extract increased during the last stages of growth, but fluctuated and was usually higher in 1891. and the same is true of the carbohydrates.1 BOLLS. FERTILIZING CONSTITUENTS. The table below shows analyses of the mineral matter of 6 samples of bolls containing- lint and seed and of 3 samples of empty bolls. The latter constitute about 14 per cent of the entire plant: Fertilizing constituents of cotton bolls. No. Date of analy- sis. Water. Ash. Nitro- gen. Phos- i phoric Potash, acid. Lime. Mag- nesia. 1 2 3 4 5 6 Bolls collected J tily 22 Bolls collected Aug. 1 . . . Bolls collected Aug. 11 Bolls collected Aug. 22 Bolls collected Aug. 7 Bolls collected Aug. 17 1890 1890 1890 1890 1891 1891 Per ct. a 10 a 10 a 10 a 10 a 10 a 10 Per ct. 6.02 5.71 5.25 5.11 3.87 3.43 Per ct. 3.11 2.99 2.53 3.03 1.87 1.70 Per ct. 1.13 1.08 1.04 .98 .82 .71 Per ct. 2.08 2.14 2.09 1.93 1.41 1.21 Per ct. 0.74 .61 .41 .51 .43 .37 Per ct. 0.44 .45 .40 .41 .38 .35 4.90 2.54 .96 .89 .40 .17 1.81 1.85 2.90 3.23 .51 | .40 1874 1889 1890 9.47 14.36 7 12.96 7.65 7.03 1.03 1.36 .87 3.54 | .79 8 do 1.09 ' .29 q do .77 .21 11.92 9.21 1.08 .48 2.66 1. 80 . 43 a Assumed. The analyses here, as in previous cases, seem to indicate a decrease in fertilizing constituents as the period of growth advances. 'Mississippi Sta. Tech., Bui. 1. 90 THE COTTON PLANT. PROXIMATE CONSTITUENTS. Hutchinson and Patterson have reported 6 proximate analyses of this product, as shown in the following' table. Analysis No. 4 is of nearly mature bolls. The others are presumably of immature bolls in which the seed and lint are undeveloped. Analyses at different stages of growth during two years are given to show the development of the plant: Proximate constituents of cotton holla. No. Date of analy- sis. Water. Ash. Protein. Fiber Xitrogen- free extract. Pat. 1 2 3 4 5 6 Bolls collected July 22 Bolls collected Aug. 1 Bolls collected Aug. 11 Bolls collected Aug. 22 Bolls collected Aug. 7 Bolls collected Aug. 17 1890 1890 1890 1890 1891 1891 Per cent. a 10 a 10 3 10 a 10 a 10 a 10 Per cent. 6.02 5.71 5.25 5.11 3.87 3.43 Per cent. 19.48 18.66 15.83 18.98 11. 73 10.65 Per cent. 14.92 23 29.73 30.36 7.04 13.27 Per cent. 46.82 38 78 35.65 26.85 64.60 59.78 Per cent 2.73 3.85 3.54 8.70 2.76 2.87 6.02 3.43 4.90 19.48 10.65 15.89 30.36 7.04 19.72 64.60 26.85 45.42 8.70 2.73 10 4.07 11.92 7.34 6.96 32.50 39.90 1 1-38 aAssumed. The crude liber in the bolls increased very rapidly as they matured, though it was much higher during the first season. The protein, which fluctuated very little dur- ing the first year, was very much higher than it was the second year, and in the last season there was a decline during the last period. The ether extract, which was very constant in quantity during the earlier periods of development of the bolls, increased greatly during the last period in 1890, when all the bolls were nearly matured. The carbohydrates, which were very much higher during the last year, decreased rapidly as the bolls matured.1 LINT. FERTILIZING CONSTITUENTS. Lint constitutes 10.50 per cent of the mature plant. A number of analyses with reference to fertilizing constituents have been made of this material, the results of which are embodied in the following table: Fertilizing constituents of cotton lint. No. Kind of lint. Sea Island. Upland .do. ....do Sea Island . ....do Short staple. Upland do do (Minimum . 10 analyses < Maximum . (.Average . . Date of analy- sis. 1874 1857 1857 1857 1857 1857 1874 1889 1890 6.72 6.77 4.72 6.77 6.07 Ash. Per ct. 1 1. 14 .93 1.25 1.31 1.50 1.50 1.75 1.50 1.80 .93 1.80 i.37 Nitro- gen. Per ct. 0.54 .28 .20 Phos- phoric acid. Per ct. 0.09 .04 .11 .07 .16 .17 .06 .18 .07 .05 .05 .18 .10 Per ct. 0.48 .47 .28 .44 .28 .36 .44 .37 .64 .85 Per ct. 0.11 .07 .16 .21 .18 .26 .11 .48 .16 .15 .07 .48 .19 Mag- nesia. Per ct. 0.03 .10 .03 .12 .06 .02 .03 .17 .11 .16 .02 .17 .08 Mississippi Sta. Tech. Bui. 1. CHEMISTRY OF COTTON. 91 This table makes it clear that if the lint were the only part of the plant removed from the land on which it is grown, cotton would be one of the least exhaustive of farm crops. The only other part which need be permanently lost to the soil is the oil, which also contains very small amounts of fertilizing constituents. Bowman has reported1 that associated with cotton fiber u there are small quantities of nitrogen, on the average about 0.0345 per cent, but differing in different varieties of cotton. The nitrogen appears to form part of the albuminous matter which was found by Schunck to be con- tained in the fiber [see p. 92]. * * * In some cases it may also arise from or be increased by the existence of small quantities of nitrates associated with other mineral constituents." 2 According to Calvert, 3 cotton samples from different countries con- tain the following percentages of phosphoric acid soluble in water: Egypt, 0.055; New Orleans, 0.049; Bengal, 0.055; Surat, 0.027; Car- thagena, 0.035 to 0.050, and Cyprus, 0.050.4 PROXIMATE CONSTITUENTS. The proximate constituents of cotton lint were found by an analysis made at the Tennessee Station to be as follows: Water, 6.74 per cent; ash, 1.65; protein, 1.5; fiber, 83.71; nitrogen-free extract, 5.79, and fat, 0.61, showing that in its crude state, at least, it is far from being the pure cellulose it is often stated to be, a fact abundantly proven by the investigations reviewed on the following pages. MISCELLANEOUS CHEMICAL STUDIES OF COTTON FIBER. Structure of cotton fiber. — O'Neill,5 by treating cotton fiber with Schweitzer's reagent, succeeded in separating four different constitu- ents: (1) The outside membrane, which did not dissolve in the reagent; (2) the real cellulose beneath, which dissolved, first swelling enor- mously and dilating the outside membrane; (3) spiral fibers appar- ently situated in or close to the outside membrane, not readily soluble in the copper solution, and (4) an insoluble substance occupying the cone of the cotton hair. He also made a study of the substance found associated with the cellulose sheath in cotton fiber, to which Schunck has given the name of cotton wax.6 The composition of this wax appears to differ slightly with the different kinds of cotton, but an average analysis of wax from American fiber is reported as follows: Carbon 80.38 per cent, hydrogen 1 4.51 per cent, and oxygen 5.11 per cent. 1 Structure of the Cotton Fibre, p. 70. 2 The figure here given for nitrogen by Bowman, it will be noticed, is much lower than that given in the table above; so much so that it appears possible that his decimal point may be in the wrong place. 3Compt. Rend., 65 (1867), p. 1150; Jour, prakt. Chem., 1869, II, p. 122. 4 For detailed analyses of the ash of cleaned and uncleaned Sea Island cotton, by Ure, see F. H. Bowman, Structure of the Cotton Fibre, 1882, p. 67. 5 Calico Printing, Bleaching, and Dyeing, Vol. II, p. 2; F. H. Bowman, Structure of the Cotton Fibre, 1882, p. 50. 6 Mem. Manchester Lit. and Phil. Soc, Vol. IV, 3d ser., p. 95. 92 THE COTTON PLANT. The wax fuses at 186.8° F. and solidifies at 179.6° F. The wax from Dhollerah cotton differs from that derived from tfie American fiber in that it does not solidify until it has reached a temperature of 177.8° F. "Along- with this wax is also a fatty acid, which is white and solid, and which by analysis has been proven to be identical with margaric acid." E. Schunck1 has shown by careful investigation that cotton, instead of being a pure cellulose, "contains a number of other ingredients, some of which occur so constantly that they may be considered essen- tial constituents of cotton viewed as a vegetable product," and which may amount to 13 per cent or even more. This fact had previously been pointed out by Persoz,2 who states that the fiber of cotton, hemp, linen, etc., contains (1) a certain quantity of coloring matter, (2) a peculiar resin insoluble in water and soluble with difficulty in alkalies, (3) a small quantity of fatty matter, and (4) inorganic saline matters. Schunck succeeded in separating the following substances: (1) A species of vegetable wax, (2) a fatty acid, (3) two kinds of coloring matter, (1) peptic acid, and (5) a trace of albuminous matter. Of the two coloring matters •" which Schunck succeeded in separating from Nan- kin cotton, one was easily soluble in alcohol and was obtained on evaporating the solution as a dark-brown, shiny, transparent residue ; the other was almost insoluble in cold alcohol, but dissolved in boiling alcohol, and was deposited on the solution cooling, in the form of a light-brown powder. Their properties are in general the same as those of the analogous coloring matters from ordinary cotton. Their com- position was as follows: • A. B. Coloring ! Coloring matter solu- matter insolu- ble in cold | ble in cold alcohol. alcohol. c Per cent. 58.22 5.42 3.73 32.63 Per cent. 57.70 5.60 4.99 31.71 H N 0 Total 100 100 The composition of the analogous coloring matters from American cotton, accord- ing to previous determinations, was as follows: A. B. c Per cent. 58.42 5.85 5.26 30.47 Per cent. 58.36 5.71 7.60 28.33 H N 0 Total 100 100 The difference in composition, in the first case at least, is not greater than may be expected with substances of the purity of which, in consequence of tbeir not occur- ring in a crystallized state, one can never be perfectly sure. On the whole, I think these experiments justify the conclusion at which I have arrived, viz, that the color of Nankin cotton is due to the presence of bodies which are very similar to, if not ^hem. News (Amer. ed.), 2 (1868), p. 232. 2Traite de l'lmpression des Tissna 3Chem. News, 29 (1874), p. 5. CHEMISTRY OP COTTON. 93 identical with, those which cause the much fainter tints of the ordinary kinds. They show, too, that the substances accompanying the cellulose (whether clothing the fibers or contained in their interior) are the same with this variety of cotton as with all those previously examined. Water content of cotton fiber. — F. H. Bowman1 states that the quau tity of water in cotton fibers "-varies with different seasous from 1 to about 4 per cent in the new crop, and rather less as the season advances. Above 2 per cent of moisture, however, seems to be an excessive quan- tity even in a new crop cotton, and when more than this is present it is either the result of a wet season and the cotton has been packed before drying, or else it has been artificially added." In testing the quantity of so-called water of hydration, the author found that on heating to 212° F. samples of cotton lost from 5 to 7 per cent in weight, "and when they were replaced in the same room for some days they gradually regained all the weight they had lost." Sand and mineral matter in different classes of cotton fiber. — TJre reports determinations of the amount of sand and mineral matter in 12 different classes of cotton. "The samples were taken out of bales upon their arrival in Liverpool." The results varied from 1.15 (rough Peruvian) to 6.22 (Dhollerah) per cent, with an average of 2.51 per cent, the American cotton containing 1.52 per cent of mineral matter. As a rule, it may be taken for granted that an excess of ash much above 1 per cent arises from the presence of impurities. SEED. Sea Island and Egyptian seeds are both entirely freed from lint by ginning, but with upland cotton seed the lint still adhering to the seed after it has passed through the gin amounts to about 10 per cent of the total weight of the seed. According to the Tenth Census, the ginned seed yields at the oil mills the following products : Kernels, 50 per cent, yielding — Per cent. Oil 25 Meal 75 100 Hulls, 50 per cent, yielding — , Linters 2. 2 Hulls _... 97.8 100 In the whole seed : Meal 37. 5 Oil 12.5 Hulls 48.9 Linters 1.1 100 structure of the Cotton Fibre, 1882, p. 62. 94 THE COTTON PLANT. While the figures in these tables indicate how the seed was divided by the oil mills at the time that the data were collected, they do not represent the actual weights of the different parts of the seed. The lint, as we have stated, was found at the South Carolina Station to be 10 per cent by weight of the ginned seed. The North Carolina Station1 found from a number of tests that the proportion of hulls to kernels was as follows: Hulls, 49.9 per cent; kernels, 50.1 per cent. The Texas Station2 found the proportion as follows: Hulls, 45.2 per cent; kernels, 54.8. J. H. Cooper reports:3 Hulls, 42.25; kernels, 57.7. Adriane4 states that Egyptian seed shows 37.45 per cent hulls to 62.55 per cent kernels. These figures are in each case the averages of a number of tests where the kernels and hulls were carefully separated by hand, and in each case the lint is included with the hulls. By averaging these results we obtain the following table, which represents very nearly the actual weights of the different parts of the seed: Kernels, 54.22 per cent, yielding — Per cent. Oil 36.88 Meal 63.12 100 Hulls, 45.78 per cent, yielding — Linters 27.95 Hulls 72.05 100 In the whole seed : Meal Oil 34.22 20 Hulls 35.78 Linters 10 100 FERTILIZING CONSTITUENTS. The seed constitutes 23 per cent of the weight of the entire plant. The following table gives a summary of the results of 15 analyses of this part of the plant with reference to fertilizing constituents : Fertilizing constituents of cotton seed. Minimum. Maximum Average - ■ Wa- ter. Perct. 7.04 9.51 8.42 Ash. Nitro Per ct Per ct. 2.80 4.96 3.78 1.90 5.17 3.13 Phos- phoric acid. Pot- ash. Soda. Perct. Perct Perct. 0. 76 0. 73 1. 77 1. 63 1.27 1.17 0.02 .50 .20 Lime, Perct 0.11 1.15 .25 Mag- nesia. Per ct 0.40 .79 .55 Sul- phu- ric acid. Perct 0.01 .27 .12 Fer- ric oxid. Perct 0.02 .13 .07 Chlo- rin. Insol- uble mat- ter. Perct. Perct. 0.02 0.01 .18 .12 .05 .06 1 North Carolina Sta. Rpt. 1882, p. 93. -Texas Sta. Bui. 2. ^Southern Cultivator, Vol. Ill, p. 111. ^Chern. News, Jan., 1865. CHEMISTRY OF COTTON. 95 PROXIMATE CONSTITUENTS. In the following table is given a summary of the results of 25 proxi- mate anaylses of the whole cotton seed, together with similar sum- maries by Dietrich and Konig, Wolff, and Kiihn : Proximate constituents of cotton seed. Dietrich and Konig's summary Minimum Maximum Average (8 analyses) Wolff's average Kiihn s summary: Minimum Maximum Average Summary of all analyses : Minimum Maximum.- .. Average (25 analyses) Water. Ash. Per cent. 8 11.42 9.76 11.40 Percent. 2.89 8 4.86 4.30 | 7.70 8.90 8.10 7.50 8 17.51 9.92 2.89 8 4.74 Protein. Fiber. Percent. 13.62 29.70 19.56 19.90 22.75 22.80 22.80 13.62 29.70 19.38 Per cent. 18.93 32.40 23. 46 18.90 Nitrogen - free extract. Per cen t. 7.58 36. 70 22.45 20.20 16 7.60 24.70 15.40 20.30 11.50 17.60 7.58 32.40 36.70 22.57 23.94 Percent. 10.40 29.34 19.91 25.30 29.30 30. 30 29.80 10.40 29.34 19.45 MISCELLANEOUS CHEMICAL STUDIES. Proteids of cotton seed. — T. B. Osborne and C. L. Voorhees have reported1 an exhaustive study of these substances. The cotton seed used was freed of the husk and the fat removed with benzin. It was then extracted with water, with 10 and 20 per cent sodium chlorid solution and with 0.2 per cent potash water. The results were not altogether satisfactory to the authors, unusual difficulties being encoun- tered in filtering the extracts and in separating the coloring matters. With brine a globulin was extracted agreeing in composition and in general properties with the vitellin obtained from the seeds of wheat, maize, hemp, castor bean, squash, and flax, to which the name edestiu is given. The largest amount of this found in the oil-free meal was 15.83 per cent, and it contained 42.3 per cent of the total nitrogen in the meal. The proteid matter dissolved by water consisted almost wholly of proteose, amounting to about 0.75 per cent of the oil free meal. The potash extract contained so much gummy matter that it was filtered with difficulty, and no preparation was made. The nitrogen in the extract represented 44.3 per cent of the total amount in the meal. The residue from the extraction with potash water contained nitrogen equivalent to 11.4 per cent of the total amount in the meal. Sugar in cotton seed. — The investigations of Kitthausen2 and Bohm3 showed the presence in cotton seed of considerable amounts of sugar. By extraction of fine-ground cotton-seed cake with warm 80 per cent Connecticut State Sta. Rpt. 1893, pp. 211-217. 2 Jour. prak. Chem., 29 (1884), p. 351. 3Sitzber. Ges. Beford. ges. Xaturwisseusch., Marburg, 1883, No. 1, p. 24; Jour, prak. Chem., 30 (1884), p. 37. 96 THE COTTON PLANT. alcohol, the former obtained about 3 per cent of crystallizable material. A study of the physical and chemical properties of this substance led to the conclusion that it was identical with the melitose of Berthelot,1 a substance which had hitherto been obtained only from eucalyptus manna.2 Bohm proposed the name gossypose for the cotton-seed sugar, and both lie and Ritthausen evidently considered it a simple and definite compound., More recent investigations, however, by Berthelot3 have tended to show that the melitose of both eucalyptus manna and of cotton seed is a combination4 of raffinoseand an amorphous unfermentable substance to which Berthelot gave the name eucalyn. These conclusions regard- ing the compound nature of melitose have been confirmed by the investigations of Tollens and Rischbiet.5 Nitrogen bases of cotton seed. — In furtber investigations of the con- stituents of cotton seed, Ritthausen and F. WegerG found that "the mother liquid of melitose from cotton seed dissolved in 90 per cent spirits gave on the addition of PtCl4 a crystalline precipitate," which was found to be a compound of betain, but it appears doubtful whether betain " exists as such in the seed, or whether it is created by the decomposing influences of acids during the different evaporations." Cholin was also obtained from cotton seed by Bohm.7 M. Maxwell,8 in continuation of work by Bohm on cholin, of Ritt hausen and Weger on betain, and of Gaehtgens on the toxic properties of neurin,9 undertook to determine the extent to which cholin and betain are present in cotton seed or cotton-seed meal. In a sample of cotton-seed meal he found the following relative proportions: Cholin, 17.5 per cent; betain, 82.5 per cent. The behavior of these substances toward different chemical reagents is described. COTTON-SEED PRODUCTS. Cotton seed furnishes a variety of valuable commercial products. The following diagram (p. 97), prepared by Grirnshaw10 on the basis of the actual results at oil mills, indicates how a ton of cotton seed is utilized. "'Ann. Cbim. et Phys., ser. 3, 46 (1856), p. 66. -See Thomson, Organic Chemistry of Vegetables, 1842; Mudie, Jour. Pharm., ser. 2, 18 (1832), p. 705; F. von Miiller, Eucalyptographia, London and Melbourne, 1885, 10th decade, Eucalyptus viminalis; and especially Johnston, Lond. Ebinb. and Dubl. Phil. Mag., 23 (1843), p. 14; Jour. prak. Chem., 29 (1843), p. 485. rCompt, Rend., 103 (1886), p. 533. "Scheibler, in Ber. deut. chem. Ges., 22 (1889), p. 3121, states that the so-called encalyn is merely an impurity and inversion product. 6Liebig's Ann., 232 (1885), pp. 169, 172; Ber. deut. chem. Ges., 18 (1885), p. 26. 6 Jour, prakt. Chem., 30 (1884), p. 32. 7Address before the Scientific Society of Marburg, 1881. 8Amer. Chem. Jour., 13 (1891), p. 469. 9Dorpat. med. Ztschr., 1 (1870), p. 161. 10 Jour. Frank. Inst., 1889, p. 191. CHEMISTRY OF COTTON. Products from a ton of cotton seed. Cotton seed, 2,000 pounds. 97 Meats, 1,089 pounds. Linters, 20 pounds. Hulls. 891 pounds. Cake, 800 pounds. Meal. Fiber. (Feeding stuff. Fertilizer.) Crude oil, 289 pounds. Summer yellow. (High-grade paper.) (Winter yellow Cotton-seed stearin.) Soap stock. Soaps. Fuel. Bran. (Cattle food.) Salad oil. Summer white. Lard. Fertilizer. Cottolene. Miners' oil. Soap. The above diagram was prepared several years ago. Kecently the piocesses of manufacture have been so improved that over 300 pounds (40 to 45 gallons) of oil can be obtained from each ton of seed, and delinting machines have been introduced which remove a much larger amount of linters than is given in this diagram, the proportion of hulls being correspondingly reduced. COTTON-SEED HULLS. Cotton-seed hulls, as we have already seen, constitute 45.78 per cent by weight of the ginned seed. They constitute the hard, outer cover- ing of the seed and are composed principally of woody matter arranged, according to Gebek,1 in five layers of cells. Gebek gives a description of the appearance of a section of cotton-seed hull under the microscope, and refers to a detailed discussion of this material by Bretfeld.2 The table of analyses on page 98 shows that the hulls tire principally crude fiber and nitrogen-free extract, these with water constituting more than 90 per cent of the entire substance. Fertilizing constituents. — In the next table will be found a summary of the results of 8 analyses of the mineral matter of cotton-seed hulls. JLandw. Vers. Stat., 42 (1893), p. 279. 2 Jour. Landw., 1887, p. 29. 1993— No. 33 7 98 THE COTTON PLANT. Fertilizing constituents of cottonseed hulls. "Water. Ash. Nitro- gen. Phos- phoric acid. Pot- ash. Soda. Lime. Mag- nesia. Sul- phuric acid. Ferric oxid. Insol- uble mat- ter. Per ct. 8.76 11.45 10.17 Per ct. 2.07 2. 99 2.40 Per ct. 0.35 .9G .69 Per ct. 0. 09 .56 .25 Per ct. 0.36 1.32 1.02 Per ct. 0.01 .02 .02 Per ct. 0.13 1.09 .18 Per ct. 0. 16 .35 .26 Per ct. 0.08 .09 .08 Per ct. 0.02 .04 .03 I'r.ct. 0.01 . 11 .05 Cotton-hull ashes. — The cotton-seed oil mills have in the past largely used the hulls for fuel, the ashes thus produced having been exten- sively used as a fertilizer. The quality of these ashes varies greatly on account of impurities introduced, principally by the use of other fuel with the hulls. The table below gives the minimum, maximum, and average composition of cot- ton-hull ashes compiled from 1S5 analyses: Fertilizing constituents in cotton-hull ashes. Water. Per ct. 0.25 22. 30 9 Phos- phoric acid. Pot- ash. Soda. Lime. Mag- nesia. Sul- phuric acid. Ferric oxid. Car- bonic acid. Chlo- riu. Insol- uble niat- ler. Per ct. 2.37 15.37 9.08 Per ct. 7.02 44.72 23. 40 Per ct. 1.30 3.86 2.58 Per ct. 0.86 19. 35 8.85 Per ct. 2.85 17.15 9.97 Per ct. 2.41 2.71 2.56 Per ct. 0.92 4.93 2.07 Per ct. 9.56 11. 59 10.57 Per ct. 0.21 3 1.60 Pr.ct. 0 96 43. 30 14.04 Proximate constituents. — In the following table will be found a sum- mary of the results of 22 proximate analyses of the hulls: Proximate constituents of cotton hulls. Water. ! Ash. Protein. Fiber. Nitrogen- free extract. Fat. Per cent. 7.25 16. 73 11.36 Per ceii t. 1.65 4.43 2.73 Per cent. 2.78 5.37 4.18 Per cent. 35.75 66.95 45.32 Per een t. 12.41 41.24 34.19 Per cent. 0.75 5.41 2.22 COTTON-SEED BRAN AND COTTON-SEED FEED. Cotton-seed bran and cotton-seed feed are products very similar to cotton-seed hulls. In fact, cotton-seed feed is a mixture of cotton-seed hulls and cotton-seed meal ground together. Cotton-seed bran usually consists of ground hulls, together with an admixture of meats derived for the most part from the waste of cottonseed oil mills, the waste con- sisting principally of immature or frosted seed. CHEMISTRY OF COTTON. 99 The following table gives a summary of the results of 8 proximate analyses of these two products grouped together : Proximate constituents of cotton-seed bran and feed. Water. Ash. Protein. Fiber. Nitrogen- free extract. Per cent. 30.67 47.33 39.22 Fat. Per cent. Per cent. 8.86 1 2.18 13.07 i 4.92 11.66 3.06 Per cent. 6.37 24.13 12.01 Per cent. 21.39 43.28 30.99 Per cent. 1.33 5 47 3 06 COTTON-SEED KERNELS. By cotton-seed kernels we mean the inner portion of the seed, vari- ously called kernels, meats, hulled seed, or peeled seed. Fertilizing constituents. — There are apparently but few analyses of the mineral constituents of the kernels. Calvert1 reports a partial analy- sis. One hundred parts of the kernels gave 3.52 per cent of ash, containing — Per cent. Phosphate of magnesia 0. 652 Phosphate of iron 053 Alkaline phosphate 387 Other salts 2. 428 The author states that the kernels contain phosphoric acid in both the soluble and insoluble forms. The North Carolina Station has reported2 a more complete analysis of the mineral constituents of the kernels, as follows: Fertilising constituents in cotton-seed kernels. Per cent. Moisture 6. 27 Crude ash 4. 03 Nitrogen 4. 98 Phosphoric acid 1. 73 Potash 1. 14 Lime 16 Magnesia 78 Ferric oxid 03 Sulphuric acid 12 Chlorin 01 Silicic acid 05 These two analyses apparently constitute the analytical data for mineral matter of cotton-seed kernels. ^ompt. Rend., 65 (1867), p. 1150; Jour, prakt. Chem., 1869, II, p. 122. 2 North Carolina Sta. Rpt. 1882, p. 97. 100 THE COTTON PLANT. Proximate constituent*. — The table below gives the results of 7 prox- imate analyses of this product: Proximate constituents of cotton-seed kernels. No. Date of analy- sis. Water. Ash. Protein. Fiber. Nitrogen- free extract. Fat. 1 2 3 Character of seed unknown . Egyptian seed 1856 1870 1870 Per cent. 6.57 7.54 8.12 7.90 Per cent. 8.91 8.60 9.44 5.00 Per cent. 31.86 27. 20 28.12 29.40 Per cent. 7.30 32 33 1.90 Per cent. 14.82 71 74 17.96 Per ct. 31.28 23.95 20.58 4 37.84 7.53 6.14 6.27 6.04 7.99 7.02 4.03 5.41 29.14 4.68 26.33 24.33 Character of seed unknown. Kernels of cotton seed Kernels of cotton seed 1870 1882 1889 5 6 7 33. 57 29. 25 33.06 7.16 4.38 3.09 9.11 19.52 15.81 37.00 36.55 36.59 6.04 8.12 6.94 4.03 9.44 6.92 27.20 33.57 30.35 1.90 7.30 4.76 9.11 19.52 21.39 20.58 37.84 29.64 COTTON-SEED CAKE. Renouard * states that in 1881 three kinds of cotton-seed cake were recognized in France: (1) Linty (cotonneaux), (2) crude (brut), and (3) refined (epure). The linty cake, so called because it contains much waste cotton, is used only for manure. Its color is dark brown. The better grades of this cake are distinguished as cotton from Catania, the poorer kinds as cotton from Syria. The average composition of these is as follows : Composition of Catanian and Syrian cotton cake. Water Oil Organic matter . Ash Nitrogen Phosphoric acid Catanian. Per cent. 8.40 5.20 79.81 6.59 100 3.23 2.02 Syrian. Per cent. 7.40 6.92 80.33 5.28 100 2.86 1.12 The crude cake possesses, when fresh, a greenish color, which on storage passes over to a brown. It contains large quantities of hard, black fragments of the hulls. Its nitrogen content is higher than that of the linty variety. It is used exclusively as a cattle food. In trade it is usually distinguished as cotton from the Levant or Alexandria. Its composition is as follows: Water, 10.98 per cent; oil, 6.09; organic substance, 77.03; ash, 6; nitrogen, 4.03, and phosphoric acid, 2.07. The refined cottonseed cake, manufactured chiefly in Marseilles, is distinguished from the former kinds by its lack of coarse fragments of 1 Ann. Agron., 7 (1881), p. 511. CHEMISTRY OF COTTON. 101 hulls. It is of a yellowish color, broken by numerous dark streaks. It contains water, 11.26 per cent; oil, 4.80; organic substance, 78.76; ash, 5.28; nitrogen, 4.43; and phosphoric acid, 1.96. In this country only two kinds of cake are recognized — (1) undecorti- cated cotton-seed cake, made from the whole seed without removal of the hull, and (2) decorticated cotton seed cake, which is simply the unground cake from the manufacture of cotton oil. The former is not manufactured to any extent in this country at the present time. Undecorticated cotton-seed cake. — The maximum, minimum, and aver- age proximate composition of undecorticated cake, compiled from all available analyses, is shown in the following table, together with the summaries published by Dietrich and Konig, Wolff, and Kiihn: Proximate composition of undecorticated cotton-seed cake. Dietrich and Konig's summary (46 an alysea) : Minimum Maximum Average Wolff's summary: Cotton-seed meal Cotton-seed meal, cleaned Kiihn's summary : Minimum Maximum Average Summary of all analyses (62) : Minimum Maximum Average Water. Per cent. 7.55 14. 50 11.86 10.60 9.80 6.60 14.20 10 7.55 14.50 11.64 Ash. Protein. Per cent. Per cent. 5.03 13.70 12.51 33.69 6.38 24.25 7.20 24.70 6.80 28.30 18.20 28.30 6.80 23.50 4.33 13.70 12. 51 33. 69 6.26 24.08 Nitrogen- Fiber, free extract. Per cent. 5.29 25.59 20. 95 24.90 18.40 5.29 27.17 20.68 Per cent. 24.20 56.78 30.74 26 29 26.50 36.70 32 24.20 56.78 31.43 Fat. Per cent. 3.46 9.02 5.82 6.60 7.70 5.10 9.80 6.60 3.20 9.02 5.91 Decorticated cotton-seed cake. — This product in the form of cake finds its principal uses in England or on the continent of Europe, very little being used in America, where it is always ground to cotton -seed meal. In composition it is, of course, practically the same as cotton-seed meal. The maximum, minimum, and average composition of this material is given by Dietrich and Konig, Wolff, and Kiihn as follows: Proximate composition of decorticated cotton-seed cake. Dietrich and Konig's summary (429 analyses) : Minimum Maximum Average of all analyses Average (28 analyses), 1870-1879 Average (401 analyses), 188) to date Average of all analyses (429) Wolffs average Kiihn's summary: Minimum Maximum Average Water. Per cent. 6.25 15.60 8.62 8.76 8.61 8.62 8.90 7.70 14.30 10 Ash. Per cent. 4.15 10.08 7.05 7.29 7 7.05 7.20 Protein. Per cent. 34.71 50.80 44.09 44.24 44.09 44.09 43.60 19.70 43.80 40.90 Per cent. 2.14 9.80 5.16 6.27 4.93 5. 16 5.70 5.40 11.40 Xitrogen- free extract. Fat. Pir cent. 13.15 28.71 20. 85 19.62 21.12 20.85 19.70 10.50 27.40 15.80 Per cent. 7.14 21.05 14.23 13.82 14.25 14.23 14.90 5.40 19. 70 16.40 102 THE COTTON PLANT. COTTON-SEED MEAL. Fertilizing constituents. — The following table gives the maximum, minimum, and average composition of cotton-seed meal with regard to fertilizing constituents as compiled from 204 analyses: Fertilizing constituents in cotton-seed meal. Minimum Maximum Average Water. Ash. Nitro- gen. Per ct. 3.23 8.08 6.79 Phos- phoric acid fst 8oda" Lime. Mag- nesia. Sul phuric acid. Ferric oxid. Per ct. 4.34 12.57 7.81 Per ct. 3.35 9.90 6.95 Perct. 1.26 4.62 2.88 Per ct. Per ct. 0. 87 0. 03 3. 32 . 73 1. 77 . 29 Per ct. 0.27 1.25 .43 Per ct. 0.48 1.26 .95 Perct. 0.07 .40 .19 Per ct. 0.12 . 19 .14 Insol- uble mat- ter. P.ct. 0.02 1.36 .27 Investigations by Kilgore and Noble1 have shown that 85.5 per cent of the potash in cotton-seed meal is soluble in water, but that oidy 2.65 per cent of the phosphoric acid is available, and that ordinary methods of wet combustion for preparing solutions for determination of phos- phoric acid in this material are not reliable. M. B. Hardin,2 in a paper "On the occurrence of metaphosphoric acid and pyrophosphoric acid in cotton-seed meal," after describing various tests and methods of estimation, says: All the reactions seem to show, beyond any reasonable doubt, the presence of both metaphosphoric and pyrophosphoric acid in the aqueous solution of the meals examined. While no quantitative determinations were attempted, the qualitative results indicated more metaphosphoric than either pyrophosphoric or orthophos- phoric acid. Whether pyrophosphoric and metaphosphoric acid exist in cotton seed or are formed during the preparation of the meal is a point worth investigating. The phosphoric acid in 11 analyses was as follows : Total phosphoric acid, 2.65; insoluble, 0.16; soluble, 1.66; reverted, 0.83 ; available, 2.49. Proximate constituents. — The following table shows the maximum, minimum, and average composition of cotton-seed meal as compiled from over 400 proximate analyses: Proximate constituents of cottonseed meal. Minimum . Maximum Average . . Water. Ash. Protein. Fiber. Nitrogen- free extract. Per cent. 5.29 18. 52 8.52 Per cent. 1.72 10. 62 7.02 Per cent. 23.27 52.88 43.26 Per cent. 1.88 15.15 5.44 Per cent. 9.13 38.68 22.31 Fat. Per cent. 2.18 20.66 13.45 COTTON-SEED OIL.3 In the early history of cotton in this country the seed was a refuse, which was either burned or thrown away. Its products are now among 'North Carolina Sta. Bui. 91d. 2 South Carolina Sta. Bui. 8 (n. ser. ). 3 See also article on handling and uses of cotton, p. 365. V CHEMISTRY OF COTTON. 103 the most important elements in our national industry. The oil is of course the main product. By far the larger portion of the oil manufactured in this country is used in the preparation of food products, principally refined lard and salad and cooking- oils. It is also used in the manufacture of soaps of various kinds, washing" powder, cosmetics, to some extent as a lubri- cant (when refined) and for illuminating purposes, in the manufacture of bolts, nuts, etc.; and generally as a substitute for olive oil. The tables of analyses given on a previous page show that the whole cotton seed contains on an average from 20 to 25 per cent of crude oil (soluble in ether), the Egyptian seed being somewhat richer in this respect than the American. A ton of American cotton seed contains on the average about 50 gallons of oil, but the oil mills have thus far not been able to secure more than 45 gallons per ton. Egyptian seed yield slightly more oil, but it is stated to be of somewhat poorer quality. The crude oil from the presses contains, among other impurities, a peculiar coloring matter, which gives it a ruby-red color, sometimes so intense as to cau.se the oil to appear nearly black. This coloring mat- ter, together with a large proportion of the other impurities ("mucilage") is removed by the process of refining. In this process the impurities in suspension are often allowed to settle and the clear supernatant oil is drawn off. To the latter from 10 to 15 per cent of caustic soda (10° to 28° Baume), according to the nature of the oil, is added and the mixture agitated at a temperature of 100° to 110° F. for 45 minutes, the precipitate being allowed to settle from 6 to 36 hours. The residues obtained are disposed of as soap stock, in the manufacture of stearin, etc. The yellow oil resulting from this process is further purified by being heated and allowed to settle again, or by filtration, and is called summer yellow oil. Winter yellow oil is made from the above material by chilling it until it partially crystallizes and separating the stearin formed (about 25 per cent) in presses similar to those used for lard.1 The latter constitutes the true cotton-seed stearin of commerce and is largely used in the preparation of butter aud lard surrogates and caudles. Another substance, improperly called cotton-seed stearin, is obtained by distilling with superheated steam the mixture of organic acids formed when a mineral acid is made to decompose the " foots" obtained during the process of refining cotton-seed oil by alkalis, and pressing out the " olein" from the distillate after cooling and solidification.2 For the preparation of the white oil of commerce the yellow oil obtained as above is shaken up with 2 to 3 per cent of fuller's earth and filtered. 1 H. W. Wiley : U. S. Dept. Agr., Division of Chemistry Bui. 13, p. 412. 2 A. Wright: Oils, Fats, Waxes, and their Manufactured Products, p. 305. 104 THE COTTON PLANT. Physical properties of the oil. — Cotton-seed oil is twenty to thirty times less fluid than water. The refined oil is of a straw or golden- yellow color, or occasionally nearly colorless, and when properly pre- pared it is of a pleasant taste. Specific gravity : According to Allen ' the specific gravity of the crude oil varies from 0.928 to 0.930, of the refned oil from 0.922 to 0.92G at 15° to 15.5° C. Wiley,2 in examinations of 19 samples of refined oil, found specific gravities ranging from 0.9132 to 0.9154, with a mean of 0.9142 at 35° C. Wiley also reports determinations of specific gravity at 90 different temperatures (10°-100° C), showing a variation of from 0.9249 at the lowest temperature to 0.8683 at the highest temperature. Allen gives comparative determinations in which the specific gravity was found to be 0.9250 at 15.5° and 0.8725 at 98°-99° C, with a mean variation of 0.000G3 for each degree of temperature. Muter3 gives the specific gravity of brown oil as 0.917G, of refined oil 0.9136 at 100° F. (37.8° C). Sutton4 places the specific gravity of cotton-seed oil at from 0.9225 to 0.9236 at 15.5° C. and 0.8684 at 99° C. Brannt5 gives the specific gravity of the oil as follows : Crude oil at 68° F. (20° C), 0.9283; at 59° F. (15° C), 0.9306; at 50° F. (10° C), 0.9343; refined oil, 0.9264 at 59° F. J. H. Long,6 in an examination of 2 samples of crude oil and of 4 samples of oil refined by different processes, obtained the following variations: Crude oil (1) 0.9325 at 3° C. to 0.9030 at 46.5°, (2) 0.9290 at 4.8° C. to 0.8994 at 49.6°; refined oil (1) 0.9291 at 4.5° C. to 0.9014 at 45.4°, (2) 0.9312 at 2° C. to 0.9006 at 47°, (3) 0.9295 at 3.8° C. to 0.9000 at 47.4°, and ( i) 0.9322 at 0.7° C. to 0.9013 at 46.5°. " The decrease in specific gravity at mean temperatures varies in the different samples between 0.00066 and 0.00068 * * * for each degree." Solidifying- point: Brannt states the solidifying point of crude oil to be 27° to 28.5° F. (—2.8° to —2° C), refined oil 30° to 32° F. (—1.1° to0°C); Allen, 1° to 4° C; Wiley, "near or below freezing;" and Muter, 34° F. (1° C). Befractive index: The mean refractive index of refined oil, as shown by an Abbe refractometer, was found by Wiley to be 1.4674 at 25° C. The variation in the refractive index is stated to be inversely as the temperature, the mean rate of variation for each degree being 0.000288. Long (vide supra), with the method of minimum deviation, using a Meyerstein spectrometer and hollow prism, obtained the following results: Crude oil, 1.4694 at 27.9° C. to 1.4624 at 46.2°; refined oil (1), 1.4744 at 19.3° C. to 1.4658 at 44.2°, (2) 1.4736 at 18.3° C. to 1.4650 at 40.2°, (3) 1.4755 at 13.8° C. to 1.4660 at 39.2°, (4) 1.4742 at 19.3° C. to 1 Commercial Organic Analysis, Vol. II, p. 112. 2U. S. Dept. Agr., Division of Chemistry Bui. 13, p. 418. 3Spon's Encyclopedia, II, p. 1470. 4 Volumetric Analysis, p. 344. 5 Animal and Vegetable Fats and Oils, etc., p. 235, 6Amer. Chem. Jour., 10 (1888), p. 395. CHEMISTRY OF COTTON. 105 1.4672 at 36.8°. The decrease in refractive index was something less than 0.0004 for each degree of rise in temperature, and was not very different from that found for olive oil by the same analyst. Crystallization and melting points of fatty acids: The crystallization point was found by Wiley to vary from 30.5° to 35.6°, with a mean of 33.5° O.j the melting point from 34.6° to 44.4°, with a mean of 39.1°. Allen gives the following figures by Hiibl: Solidifying point, 30.5° O.-j melting point, 37.7°. The high solidifying point of the fatty acids of cotton oil sharply distinguishes it from the true drying oils. The fatty acids of linseed oil solidify at 13.3° C. Eise of temperature with sulphuric acid : This varies, according to Wiley, from 80.4° to 90.2° C, with a mean of 85.4° 0. Color with sulphuric acid : The colors produced by concentrated sul- phuric acid are with crude oil very bright red before stirring, dark red to nearly black after stirring; with refined oil reddish brown before stirring, dark reddish brown after stirring. Wiley observed that with refined oil the color varied from deep reddish brown to almost black. Chemical properties. — Cotton oil and oils of similar character are classed by Allen in the "cotton-seed oil group." This group is inter- mediate between the olive oil, or nondrying group, and the linseed oil, or drying group, since the oils of this class undergo more or less drying on exposure to air. As regards drying properties, the more important vegetable drying oils stand in the following order: Linseed oil, cotton oil, rape oil, peanut oil, and olive oil. According to Braunt,1 pure cotton-seed oil consists principally of palmitin and olein, palmitin being separated at 12° C. Gebek2 states that the pure fat of cotton seed consists almost entirely of palmitin, linolein, and olein, and is for the most part neutral. The presence of a small amount of linoleic acid explains the moderate drying quality of this oil. The amount of glycerol produced by saponification of the oil is reported by Allen to be 9.5 per cent. In investigations by Stellwaag 3 4.35 per cent of lecithin was found in the ether extract and 1.52 per cent in the benzin extract of cotton-seed meal. Cotton oil properly prepared is generally neutral, but Allen states that "cotton-seed oil expressed in England from decorticated seed often contains so large a proportion of free acid that purification with alkali becomes practically impossible." He found that a sample of oil expressed from the decorticated seed in Liverpool "required 14.1 per cent KOH to neutralize free acid." Stellwaag found in the ether extract of cotton- seed meal 3.24 per cent of free fatty acids and 92.89 per cent of neutral fat, and in the benzin extract 16.31 per cent of free fatty acids and 80.98 per cent of neutral fat. The amount of free fatty acids would probably 'Animal and Vegetable Fats and Oils, etc., p. 236. 2Landw. Vers. Stat., 42 (1893), p. 278. 3Landw. Vers. Stat., 37 (1890), p. 148. 106 THE COTTON PLANT. be determined largely by the age and state of preservation of the samples of oil or meal examined, although Eeitmair has observed that cotton seed meal kept for several months in a warm laboratory showed no change as far as the saponification equivalent and iodin number of the ether extract indicated. Gebek found 0.058 per cent of phosphorus in the ether extract of cotton-seed meal. Saponification value: The results reported by Allen show the saponi- fication value of cotton oil to vary from 190.8 to 19G.8, while Wiley's average is 197.7. Wright ' gives 195 as an average. These figures indicate a close relationship between cotton oil and the vegetabe dry- ing oils, for which the saponification value varies from 187 to 190.2, a relationship which is further confirmed by the iodin numbers reported below. Stellwaag, in the investigation already referred to, found the saponification value of the ether extract of cotton-seed meal to be 194, of the benzin extract 196.4. Iodin number: Cotton oil possesses in a high degree the property of absorbing iodin. " This is due not only to the large percentage of oleic acid which it contains, but also probably to the presence of a small amount of linoleic acid or some homologue thereof. In the samples examined in no case did the iodin number fall below 100, and in one instance it rose to 116.97. The mean iodin number was 109.02." 2 The limits given by Allen, as shown by the work of a number of analysts, are 105 to 109. Wright gives 105 to 108, with an average of 106; Schadler, 106-107; and Benedikt and Lewkowitsck,3 collating from a number of sources, 102 to 117. Holde4 in investigation of the accuracy of Hiibl's method found iodin numbers varying from 110 to 115, the higher figures obtained being ascribed to more complete satura- tion of the oil with iodin by the improved method used. In studying the ether extract of cotton-seed meal Gebek 5 found, as Eeitmair had pointed out, that the iodin number varied with the purity of the extract. The extract with ordinary ether from air-dry material showed an iodin number of 96.5, while for purer extracts the iodin number was 100.8 to 102.2. The iodin numbers of the fatty acids of the oil are given by different authorities as follows: Schadler, 112 to 115; Moranski and Demski, 110.9 to 111.2; and Williams, 115.7. Keaction with silver nitrate: Another important property of cotton oil is its power of reducing silver to the metallic state under certain conditions. A test based upon this property, first proposed by Bechi and since modified by various analysts/' may be applied either to the oil 1 Fats, Oils, AVaxes, and Their Manufactured Products, p. 181. 2 Wiley, loc. cit. 3 Oils, Pats, and Waxes, p. 306. 4Mitt. konigl. tech. Versuchs., Berlin, 1891, 9, p. 81. 6Laudw. Vers. Stat., 42 (1893), p. 287. 6 Analyst, 1887, p. 170; 1888, pp. 98, 161. CHEMISTRY OF COTTON. 107 itself or to its fatty acids. The silver either forms a metallic mirror on the sides of the vessel or is reduced in the form of minute black particles, which give a brown or black appearance, and in some cases a greenish tint, to the liquid. Cotton-seed stearin : The process by which the two products which pass under this name in commerce are made have been briefly described on page 103. Brannt states that the fat separated from cotton oil by chill- ing (to 43° to 54° F.) and pressure is palmitin. The melting point of the stearin varies somewhat with the extent of pressure, but it is gen- erally pressed so that it will melt above 30° C. Allen reports the following figures obtained in an examination of the substance: Melting point, 32° 0.; specific gravity at 98° to 99° C, 0.806; saponification equivalent, 285 to 294. According to Allen, by far the greater part of the "cotton-seed stearin" of commerce is simply the product obtained by the distillation of " foots," as noted on page 103. " Products of this kind appear to contain a large amount of unsaturated solid fatty acids, possibly isoleic acid. A. H. Allen found that a ' stearin ' of this kind had the specific gravity 0.868 at 99° and melted at 40°, while the iodin number was 89.9 ; the theoretical value for pure isoleic (oleic) acid being 90. 1."1 Coloring matter in cotton oil: Crude cotton-seed oil contains about 1 per cent of a peculiar coloring matter, referred to above, which is sep- arated in the process of refining. When crude cotton oil is saponified and the resultant soap exposed to the air a fine purple or violet-blue coloration rapidly appears. This is the so-called "cotton-seed blue" which, according to Kuhlman, lias the composition CnH^O.,. " It is amorphous; readily destroyed by oxidizing agents; insoluble in water, diluted acids, and alkalis; sparingly soluble in carbon disulphid and chloroform, but more readily soluble in alcohol and in ether; and dis- solves with purple color in strong sulphuric acid."2 According to Brannt, the unoxidized coloring matter of the oil is insoluble in acids, slightly soluble in water, and freely soluble in alcohol or alkalis. In its dry state it is a light powder of pungent odor, brown color, and is strongly tinctorial. According to J. Longmore, quoted by Allen, it is of a golden-yellow color, insoluble in water, and " dyes well and per- fectly fast on both wool and silk." Under the name of gossypin it is used as a coloring matter in the industries. It is stated, however, by Gebek,3 who studied this substance, that it does not make a permanent color when used as a dye. TABLES OF ANALYSES. In the following tables, unless otherwise stated, the figures refer to upland cotton. These analyses have been gathered from many ' Wright, Joe. cit. 2 Allen, loc. cit. 3 Landw. Vers. Stat., 42 (1893), p. 287 108 THE COTTON PLANT. different sources, and where the number of analyses is large their aver- age must represent very accurately the true composition of the substance in question. This is particularly true in the case of the seed products, where we have grouped a comparatively large number of analyses. In the case of the leaves, stem, roots, and bolls, where the composition seems to depend largely on maturity of the plant, the average means very little, and a great number of analyses will of course be necessary to determine with accuracy the composition of these products at differ- ent stages of growth. We have, however, inserted the average in all tables in the latter cases, as much for the sake of uniformity as for what the figures show. It may be well to state here that in many proximate analyses the proportion of water was not given. When this was the case we have, for the sake of uniformity, assumed 10 per oent of water and calculated the analyses accordingly, the fact being stated in footnotes. Again, in several cases the analyses have failed to add to 100. In this case, when the difference from 100 was 0.5 per cent or less, correction was made in the nitrogen-free extract and the analysis included in the average. Where the error was greater than 0.5 the fact has been noted by inclosing the analyses in ( ), and the analysis excluded from the average. It has not seemed desirable to give more than the maximum, mini- mum, and average composition of cotton-hull ashes. A large number of analyses of this product are on record (185 having been compiled for the purposes of this article), but as a result of the conditions under which it is produced it is very variable in composition and individual analyses would be of little value in a discussion of the chemistry of cotton. CHEMISTRY OF COTTON. 109 6 izi .— i w co ^* io CO t* oo oa ow c^ co -rr 10 ot-ooao ~ e P3 "a DH- ct 02 ,-H OWN •* S — : c3 C ^3 r; Date of anal- ysis. 1 1 § • %%z g 0.2 « *■' ?S - a a1" S s |-§| & 2 2- a Ph 0 0 — 4 b p b — s ■J s - t if..- o fe -.- - s D t c- a 0 1 b i.= o i a. z 1 ■a '3 - ■a« §1 P- - 1 ~. DP,t — » Sr.T O i - ; It D I ; »^ OE ; i o Pi ^ b a cs * 1; c^ i * 0 b blffi - I I t a f — o 9 * - 1 t ■32 5 P-.2 1 ) 1 i [| bi = 13 oS -a" IE I 1 d b is-? at b § 2 c col'E h — 5 a &5 1 6 t- oc - 1 '3 J2 1 8 ~ t a III 1 0 2 0 <0 bi a bi e pq D eg bi a \~ a t 1 - :f.-- ^ - CD D bi c cc i . l i '0 ? - a cj 'E * [ S bt ■2 .a pi Z © o. ev -* •c cc t- OC a = c m -* m CO t- X CM CN CN CO 110 THE COTTON PLANT. d -hi in cot- coos oh CM CM CM CM CM CM CO CO B CO -1 cr CO CN Cr" Tfllf GO r4 t-i CB a O) r< 0 IB o a a u <£ CD IO "3 pq ci CD WD CD PI O O a a & ca b" o co u CO ■a 0 -De H 1-3 c - c re e e C oj . 2 2 = 2 °?^ ■ j^ cu ^4 Phc SPQ ■ a A <*> B fl-g 00 r in i • 1 1 l;« fl _! -f • ft O * r-5 t--* CO i-H Tl CO -HI rH CO O in CM CM HHCI r; in co CM CO o in co o CM CO CO •# in CO CO in CM r; CD CO ITS -^ 'f O i-l o O ' rt r-i co in o ■* — 00 O CM -* O OS -i* ^x co co in a 3 r; CMCO O CO oo iraco M O r-i -di -J EHOHJ 00 00 oo OS 00 O OS CO o ^ r_- co o c8 o CO r; HOD CM H -P ■* t-o Hf CO cs -rH co co in CO IO rH rH O Ph 8 ■w oo co o os o m »-i o « OS O tOO P^ ri c4 co i-3 co c4 coco as io t- iH CM CM rH CO CM CO r-( CM r-i in -h oo O CD Tf i-i CO CM m OS CO -rH OS i-l r* COCO CD Phos- phor- ic acid. *a *h ^ ift OS OO COO « 00 t~- GC CO 00 CI 00 t- r; ■* CO 00 CO -H^ t-CO OS Ttl -r# r-( CO in cd co CO -H 00 CO t> co r- -T* oo CO CO rH rH CD .A 6 a *a CO ^ COCO h--# lO -* W00-* 00 OS CS CO 00 OO 00 111. ° ,a ci t V a : la eo £ H O ■ ttiZ i5 rim 2 J E ° b 3- (5 cS C a b a b P* o rfl ft _, a rs b B e B a a 1 e fcjo3 .: t .- - ~ "3.8 5 ) 0 Q ) b i.l a b .= ': Pt O ,4 ft a Dd o i- 1 6 'is'- t o « ? Pt° - °E — ) D 3 b of 0 b fl c« 1 1 6 o^ ! DA'S , a K b © o c SrB | ) Q u| 1 pc fl M<3 rC q c a c C 6 6 •^ « if cc oc a c a- o- cs sr ■?_ IC CN cc -t IC CHEMISTRY OF COTTON 111 «t-oooo-HWr:-finat-oo qoqo^: 1 Tf l- r r- oo a in o oo c i— o . a «s O O o •- 8H O o H 9 5 Cr • (3 a p r So "IS A j o ■ - 5 w • fe » b "O CD H iH O ^ 3 - o o H o I J-1 -— <1 t> - 5o CD oJ © - off it: ?cd "3 is'3.'' - S* . t^*3 § «o 5 £ p»-- g y s o s ftO . ,-h . Ha J, ,Q " in O W O O b C3 j_H- d-d c^t;:ooo : croc cc c o OCOHrHCDh-CO«D-i^Wt-' ^Hvt-HClH-H-iHOHO 00C0OCQOHQ0CKH1OHH00 HHC3HHCOHHHHCOCMCN O i-l o ^Tn^os^iatoc — o ~* os c i OOOOOi-lO^-ti-li-li-lOW CO CO CO CI i-i rji -t* CC O CO O CI C 3 O X I' « :i CO O M CM i-t CO lH OO) HHPlH-irlnHHHHr-IH CO t^ O -* X - O ^ O O -t C5 CO irt O OOOOOfHOOOr-lOOiH IftHHOJ^iDCIOr-CO^OO iO -* co -* co to -* co -t : t co c i r i ifiMHtDOCOOCOO'OOCOW O CO r— CO t f rH CI O t- t- lO) •— < CM Ol ro co CO co co -f ro l— i.O T -* ro ro cn ci lO ITO O O CM CC X O CM CC CO CM rC -f CO lO (M CO CC o o co ■ CI CO ^- X >■". o - O X X X N X LOic^icr^HciOjr-xxcoc: CQCOTHHOHHOOOOOH cs -o CO t- "•# O -— i-fiO-fCOXCO.— IX O-* CO CO CO CO .— ii— 1-Hr- 1 CO i— I i-i --h -^ CO Ci O CO O O O CO tr f LO CO — < ci o w o i* h x 'f o c. c o r- |> CO CM LO W CO O CC OS O t- lA OS mcOCSiOCOCOCOiO,iOlO-#-^-f t* H lO -f C] X O O I- t* H C x -^ ■* h c: x x ic, cj x r- 1 -- nroo Xr-nnOCWC-HClCl^NiO OC0U0 ^^co^coi^^^^cococoeo 3 O 00 OS CO CO CI CO oso— 'c;m-t-f-Jci-fi>t>(N ro i-H -+ os t— co -# co e j -* c-5 o cs -*-*0 CO CO CO 0 CO ^f CO -*co ^'cdoco'cios«OLOirc)^rj3^co CO OS ■«* ooooooooooooo e_*Sj3 e g e s_e « eeee OOOOO^^Hi-Hi-C-^i-IHi-t cs cs cs cs cs os cs os cs cs cs cs cs CCCOXXXXXXXXXXX OS CO CO* O OS OOOOOOOOOOOOO CO rl © c'S'S'S^'ccr'^'creee o o o o o • CO OS- O OS CS OS OS O. CS OS OS OS ~ OS OS OS X XX OC X X X X X X X X X X X X CM CM -HTIOCQO "COX i-*Cli"-t-*ClHCloOT->J>a t>s ci bio tfi CD©C&CD3,c;CU — CO CD j >. ii txi tio p § >i >j >-. ti ti tic ooeo':'3o':';3",:o 5 5 - - - bS - = « X CS O — * CO CO -rft I I CO t- X OS O 112 THE COTTON PLANT. "b Psg2 3* rH CM CO -HHtOCO t~00©©rHCMCO-fiOCOt^CO© ofE| > a -S i W o bO t> _g ■IS* arH "Fiijy O^1? ooooooooooooo « t^^i"-~r-;T* .T3.8 S f 13 "» * * « •« tJ * tS TS 1= -o O O W hi w ft^cM rH CO O CO lO'+'f^owmfOHtDoinio © CM CM OTfKDO'tClOIMMtOfO'^'* 0,0 h o •* o c; o -* cm © hhOhOh <£1 -<# 00 (M M-*ncOSCOOHONOt*t' oo co oo oioiHxQOOMinoinos-^oo in ^ m m .h r- m m CO ro 03 CI tn la's £-g &,o CM ©t»©HNON^O(0©M« ^H HMHHH^IHOI CO.^ ■« C4 M OlflCOh-MiO !-H O O O —i CM CO (DtOO COQOOXCOCD®Ht-MCCO:i © HO HHtDmWOMONX-^HCO CO CM CO N©CONNHNCONnP^HW CO t— l> -H CO CO O CO O CM O © « CO CO Q^cm i-i 0000tr-rHl.l0CO-#COrHOS©©tM iO^NN«DrOOO)^ttinX^ OO -* OS CO rH rH OHOO^N rHrHrHrHrH(Mi bJD bJD bJD S a >> >> >s bJD cfi bJO j © © © ■" ■ bcoooocjocuoob CD©iOCD©© >s bi, bi hi bio bi 1-3 SS-4 CHEMISTRY OF COTTON. 113 "J ° OS:-'73 c-t CM £ o in in g CO 00 00 ° fc is* rC •* (SPA* -*" -* 3 00 -2 N -73 S PO 6 p, a 2a"3| «aj« « "4 ►. is o fa >>-s _; _5,q Si J3 s — «8 s a ,g O - O X O 4a O-3O«OO0Q a o«ls fag c3 ci"w ■w • += © m g as (O rH iH M r-. i-l O O m O CO H i-l O i-H t> ci ci -* co m »-i OS = PS=-=p .p 0-1 OS r j5 'tii ee 5 «« § o MPnW^t?'^ Rfc§ o P- : * gpq q t- ftO J "a! *ai -3 P. c c i, Hhl S'rj 5W § o fe P p=n P p t- Pi H . . C © "1 -* ^ ft C3 -H « OS o ? jj W ^ m r^ M '" - 1-3 -i PfcH . ft g* »? - - spw cm i-s (M Ha ci r* .,- . rq m din p.h ftO^ ■ «pq £: P< . ft . P. . r-i » 'A O O W H h -* 5 S a -S - ■§ a © r; h « Wcfl 3 K3 '■-' r« *i a) ci ci •3. g & «H 5 5 bH.,2 R»ftCO COHTf t^ HOHQO S.m-3 » o ° « £. 00 00 00 CO CO 00 00 coco co co cr. x x x r~ n t^ 1^ x x x x n oo 0000 ooxococoooo o 0 as te-4 0-3 " d o .ft -3 h q K o .a ci t> p p iHCl CO T)t in ep t> CO CHEMISTRY OF COTTON. 115 rn:Nco-i to eS § P OS _o +3 7 * *- f Z & ci ■f. s> q 3d C£ 7 c 7 R5 = cj = S •75 3d 3d © © CD X X 93 «»ajooooooi»oo+3a)00eyoosocoooooo30ocoooo« s a s^s^-- -3 s^ crs a 9 fl-w-B p'd'tf P-otS'd'^'aTS'e'a 3^-5-3 — ^rrr; a ~3 tjj — 3 33a o o c B O O a H 2 S o u o O O Ej C O s - o .96 43.30 14.04 1 .-* o CM O "co ih o 2.41 2.71 2.56 CM - - .92 4. 93 2.07 O 2.85 17.15 9. 97 00 .80 19. 35 8.85 O o eo'ao co oo ta o t- -* Th Ht-HHIOHU Q ^* ~. Q C5 X C H CI H t-i i-i H Cv 00 o CM IT a. CO -# CO-tHCOSOCMCOOCT oc oo ococrir. — t- j-i rA CM rA ci rH ci ci r-i r- to co co er 00 0C OJ ^ G>00C00)0>t~CAa000OCQC0 cm it -I CO H'tffflOHHOflJ io a o 't " co -^ c ^ o o noooa-^McaaiooOT^ooiAONHooin-^-^c t~ O L~ Cl ■* t CO H :) f X L» N c n " r^ m n Cl c o o ? oi cm ci co co co co co co cm c-i ci cm ci re ci co ci co co* ci co' co er M in ro -T rH 00 ro cc r-i OO OS CM CO .-i Ci CE ~. -* CM Ci t- rH OOCOlO^CO^W-HHCaCgOIOmQQOONt-tf O t> t- C- O L- CC t- t- O O t> 03M50'0t-t-l>300t-t-0ffl(-X00: m o c O CD •-} CO o in c CM 0" 35 CI CO i W X -«f J H E W E £ •" O E S = - c a 0 a a H la jjj a J 1- 5 i a ooooooooc < CM co «# ir: to t~ oo cr. o — ci ri-ricot-xcErHci^^i^^t-xooHcin-fio© Hrlr HnriHHnHCi:i:i:i:i:i:i CM CM cicocococccococo 116 THE COTTON PLANT. 6 co COCOr- CO CO "HI ^ CM CO 4 inat^xoiOHMmHiioiDt-oofflOr- HiHiHjTjiHiinmmioiflinioioioiniotc cm co -^i ir CO CD CD CD cd i- oo os a CD CD CD CO t^ 1- CM CO c© M u at a n0lOlHHt-C0H00«C0 ^ Oi t- O 00 CO NiOOOtOmcONHWrtCOCOWHlMQO]^ kjco* cot>cd^# >#t>co*t>i^b^^i>t^r>i>i>t^t^c^^cDt^t^ CO CD CM cm -^i in co in in t* CD 00 CM rH rH -^1 in i-h cm -»ji m m -hj< t^ CD t~ t- CD CD CD co r; - - - — - O 00 CM O00 00 CO in t- o os -^i co o oo' [^ t-' OS t> t> l> r; O CD O oo cm m t> 00 OS Date of anal- ysis. : C c c _0 : ~z e -a •i 1 a ■ t i n £ c r ■s c e c c c c c e 0 e "5 e z rz e - X e — £ c c rz c z 0 c d ft cr a- c t CN <4 -^ ■< 1C CO OC E ur^ 1£ ic in in CO 1-1 a IC e CO 5 CM CO CO 9 IT CC t^ OC CO CO ir CO CO t^ CM CO CHEMISTEY OF COTTON. 117 ^irt©N000)O^NM^WOt^C0C5O^«W^Wt0l-XC. Or-7)cO'fL';'Ot^OCCJOHNm'^if5COt>OOCC^:i: ■NOOOOWQOQOOOOOOOOOOOOOJQaaOJOOiQOJOOOOOOOOCO' • CJ CI CI CI CI C3 X F< a a 0 IE CN |K D — a s- X a 'Z T -^ e: P t- a c a r a a ? - a o PC -7 _ 7 d cj cj COO ;Mffl 815 ;ooooooo-=oooco-sooooooo-scoooo'Sococcoo 0 o 1-1 O Hi o Hi 5 Hi — s X X 7 S « r X s ■«# m w C- -rf (M cooif)cao»t>i>c0ioaoo»t~<0cecQ oooon-tooo-fLici ^fiot-CMco:o»- rt ri r: c u*; -f :i ^ oo co m as m o . i o r: -# l-j « x i-1 - c t^ cc • «HOHHMTfOiOlM Hft*0)0iOCCHnC0C0^CO(>XOOHH^00 L'-.in-tOCOWXCWOOCO'l'Xt-HgH t> CO L^ CO l*" CO l> t> CO* t>" t> t> t^ t> CO r> CO CD c^ t> t> cd CO > CC > iO CC i to b^ ^" l> « » o t> ^ iri t^ 1^ t> t^ O O *^ t^ L^ t^ o "^ t^ t^ ^ t^ C. CO [^ r-il* CKO O . X t-" OS © CO* CO CI OS t>l>-t-t>t^l^CX)C0C000C0CCC000C00005CftO C. Oi CD cr. C. 05C1OOOOOO© — ■ 3 O H H ^h H H ^* « p- r- i- CI CI M W CI N 118 THE COTTON PLANT. ° s a- M.p, cS e8 ».g ° o o2 ^ p. «a NciwMcococommmco cc co c: -+ -f -* -t -* -* -* ■•# -* -+ »n >ci lo io lt m tr. tr. i- io '^ '^ -j cc c (o t> oo r- o ci o -r ^H OS t> c- Ci O CM t- O q ^ h h h co :i x o. i- -r ci re :c re re -t co ri to IO MS : : o c ocH^c^c o o a ^ -^ CO Pa & ci ■w -f (M CD O rH c; t-- CO CM CO ' « W O O CO N 00 X Q O CO : fij W H CI H H ' ' " CI H - h « co im © o i> oo o o h w in "X *i in co o n o o o t- --+ -# C ■CMOI>CMC;[>CMt>lCH ClMCMrHi^i^rHrHi^CMOlrHCMi^rtMr^lCNJCM 'LOOOH-li- O CI cooowHHooNcoHnnMomi>ioxomocoHv^ cM?iCNci:o**wo«LOcccoHi>MHr-iincMir. xooxo'tcosmio cocoM«^coeocQwcococqcocqc4c«coNcocNiciic h- OS t— H C -f CT. «f X'Xr- i lO 00 OS OS O0 CM CM CO CM CM -3< CO co to lO CO CM CM CM CM ^-^ far. 1 OS ^H O CM j io ■** -t t-^ r OS OC CO CO CM CM OS ^-tXt-OTfCOOJ- co OO OS IO CM t— lO CO CO tO CD t^ CO* CO* ■ lO O ■+ X W O '- X' C M Tl X' CI X CO CM CO CM fc- IO Ci O iH t- OO CO W CO -* r-t IO CM 0O *0 »» ^ t> O CO O iH t- O C >?moHcooot-t"foci ^> -ji cr j t~ cm co ro t> cn ^ io co ' oi o' i> co co" x »rj to oo d h Ri»S' CO^CX)OCMCOCOCOI>IOCO HiOOlflXX OS 00 CM O CO t> lO 00 t— I CO 00 t- 00 CM CO t^ *■* t>c»i>i>co"coco^co"^ioi>co*i>co"ci6i>co c= co oo io 00 CO t> CO ■ tr- CC X X OS O- OS c o . x x x x ->: x x x os os os rs os os — : i r-< cm cm cm cm r i ?i m cm re rcc r: re re re; r: oo ** -f io to i_-s 10 x ti J OS CS OS CS OS OS OS os os — X X X X X X X 00 OS O OS . X 0C 00 X X X X X X X OC 00 X X X X CO X X X X X X X X X X X X X X 0^ X X X X X x CO cprr-oposo^eiro'^ioco t-xc co ro co '*-*-+** ■ Hoiro^mcot-xoso — :i r: -f l- o t- x o c ^ on « ^ in : io o uo »* i- io co co so co co co CHEMISTRY OF COTTON. 119 CO l> 00 05 O -< N n "* O O t^ X * O - M r^ •+ i' ^ h X C O ^ M « -t o cd r- X C~. O — < CM ~ -^ — - ) o o o o © i c j -M cm cm cm w ^ ^ :d l- t- t> t~ t- l- t- t— c— t~ x x x x x oc x x x x c: o: c-. s: ct - *. r. r. - c c c c c o ; o a P ; CD ^ X i-l x x h . t- CO t> COQQ cs ce xtx c C ^ c c 1 "3 CO c c o — c rffl CM « C3 CS X o CM CD t- O CO CM - - - Mj t- o 2 CO CO CM ocot'O^^oDO t-L^"-C'tiCHci?;c;t-iooctr''jt"i,LCi>r;cc-+t*LCCO c-i o-'i oi Co' ci ci cvi cni <-^ ci o i cn cc c*l c4 co c-i cm c*i cccci^c:Tfr-t^oor-T-Nwor;c XHxx-iH?ixn 02 o o o 3. 35 9.90 6. Oo 7.65 8.55 8.28 8.16 9.12 8.89 7.46 6.63 0.90 (8.84 7. 45 7.80 7. 79 0.17 8.55 0.89 6.52 9.36 6.98 0.70 7. 79 6.13 7.18 8.35 8.36 7.94 7.96 7.67 7.74 7.05 8.05 6.45 8.22 7.65 5. 57 8.96 7.08 8. 05 co m x -*CM l- c cr. c: ~. — cc ~. ~. ~. r; ~ ~ ~ ~. ~ t. ~ c-. cr. c-. or. c. ~. t. ~ ~ c; r; ~ n r- CM I- £ t> IQ cc b- t> — c t- X X (M X OC X X CO X X 00 00 --- ~. CM ~- ~. — ■s. 00 — r. M IN o ci ro = CM cc . 120 THE COTTON PLANT. "* m CO i-H?»ieo.**iocot-aoas©»H ■«* erccococococococ-icocorocccococo -j -fia cm t lOlOH co co -** O CO CO ift ©■fHt^Oh-iOClrtulXOiOOC'l O O O CO N CO t> 00 ©'CO © CO <■* 04 OS ^t>coiot>toioa>cdu£coeQ^eocN OS 00 © S^ OS iO * CM l> OO -"# tH r-t CO C J Ift OS 00 <-i *■+ rtiflCOHOOONt-HNNTjlMiO coeD^^eo'^oot^-^-^TjHcocococo ONTt co" t^ ^ t>ooooooo ooooooo OSOOOOOO^»-lrHT-HrHi-H'-*W OOOSOSOSOSOSCSOSOSQSOSOSGSOSOS oooococoooooooooooaoooooooaoQO CO > CO CM j 0J 1 bibibjb CM -M ^ CM CO c ; CQ iH CM ^ ,H CM "^ CM CO r-4 CM « >*>> >s bio bi bi g Sj^J^fc g O C 1 -J O ■. Jj S S i K S S Jl ^ooooe'oo'o'o'oooo'o HCim^iociocoajOHiMw-^in 33, 2 3 O g S ■m ca p. CHEMISTRY OF COTTON. 121 CI CO -<* in Ot>QOOSOHMW 'H* m CO leirt-innci o s- sa ° a 7. — CD ■Sjd • h. S ° = K tig -a ■§•§ H ® . H S OQ a ffl a 2*« s. 0 — :- =- D «i 3'5 .2^.2 g-2 : :.s§.S ' ' ■ • a a HS O « t^ 1< Ol ift W COtOOO HHHH NNHHNHCQfN Ht ■**< t-~ i> cs winoii>«it-Hio owt-oo Nt-iaN^c -—< o o m o 05 m co «d a -h co Tfifiift CO CO CO CM COWNWCOCOCOCO co co co MtDOfflOiHLOt-fflOJ^OOOO'fN lococoacocoooococu^^'xioo cp^^^io^t>co"in*(dco*t^eocooo" NHO^«mt-«o*t- it-^MpJHOOHOiOii co-fco-+csmt— cot-t-t— iot*x-t ■^cococowcococococoMcjroco'* ClOOO CO CO CO t- «-H in t- O -f M © -t -I lO H *** -H CO CO CO CO r-i CO^OlCOCOOO Ift t> i-H ONHQO t- OS CM lT~ 'HH CO OS CO iMiNlft T* •<& -*# -* ^ -^ co co co co co co ># m rp ■•* 05 o -* co m - 1 in cs 1— ( as cs c o©Ht>t*HOHi*coir,Nt-osin O CO C0* iH CO* CO CO* <-h i-5 i-H cm* iH cm i-H OS m ^h m OS* OD CM* cs -* m cm co o o co t^ioo ^o-voHtomH co cm cs O © OS o »©OHHfflOO Mt-( ^ rHrH COCOCO(r--H-COCO"^OSOSt>CO'*00^ ot^int^ooO'tOHiNHooiat- in03"-rCOCMOCSCO-HHCOCMCMOaOCO rHCMCMCMCMCM»-HCNCMC^(MCMCM^-rH CO OS O rH 1 -t »h oa «■# t> o cm as OSC^CO^COCMTftCM cocNcs'co'inin-^'^ji ^CCOOSCMcOCOCMtr~CSt--COCOCOCO CMCOOOSCOCOOOS-— coint^aoost- < co ^ ^ in cm co * I CM .-H 1-1 CO i-H • 0000 00000000 000 000000000000000 OOOO o o • i CS OS OSOSOSOSOSOSOSOS OS OS CO CO CO CO CO CO CO CO 00 CO CO CO CO COCOCO OS o o o o O O f cccsoscscscsosoc.csc cr. cs as cs -ICOCCCOCOCOCCCOOOOO-XOOCCOOCC si si £ 2 _>> >._>> si CD © CD © CD CD CD a o o cd o o"S >-. s*^. si sisi£2>i>. >isb si si S3 5 5 2 2 3 3SS52 2 2 a -a -a 'OS CD CD CD CD CD CD CD CD CD CD CD CDOCDCDCDCDCDCD CDCD ejooo oo a o o o o s o o CD CD Oi CD CCCDCDCDCDCDCDCD CDCD El 2 5 sc aa«4 laa-^aaa^aa^s ™ ,&■ >i— • ^Saq ir-ir-i? ij y '* Prt •a'a-a-a-a'a'a'e'e'a'a'a'a CDcDCDCDCDCDOO, 1. CD CD CD CD CD & . CD CD C CD CD S CD CD CD CD CD CD CD CD CD S a 3 2 a M = = a a'S 3 .S cS f* ©©aJCJCi-CSiDCD jS'o'o'c's'c c oo"o':'oc'oo 300000000000000 cm co ^* m CD I— CO OS O ^H CM CO -t m CO riClrt'tlOCOOOaOHlNCO-tiCi 122 THE COTTON PLANT. CI CO -^ in CO o 00 CM CO ^ ITS CO t- 00 OS o rH CM £ o >. +i S tc {a c >. AH -1 c . CD 9 oo ^ H s 3 K CO cr A 06 K- 'S to a h »H g cu 2 ^ CD' 73 ^ ic 5-5 3 g CD c3 u CD . u O s irf'P c i^d2 HcS H-j ^ tz cc - rH r^. a a i a . CD CO CD S3 M t- in 0 CD CD CD pq > CD p CD? 3 V r^ CD CD M CD H 03 02 IS r bi "3 :o M a's * Pi Ocf «3r& D- ll"^ '& R ? O r- 1-5 - — 'r— CI > - CI 2 o o o o c a •Sooooooc !•§ w 2 Otf O O re's ^Th-Ot: e ^■S'd'S'S'O'St .8 p a CO.— o— < £1!^ a E- 5 Ch 00 <1 ^coin^ocDt^ not. a ^ ■fHQXOOTC o -^ -« o oo g C- 00 o t- r- oc t- t^ c cr 1 tt mr-^ooct-mT ■^ CO OS I-1 -1 CM rH ca §j C4 CO CO 00 CM CM CM 00 -r* 1 cs t- o> i- -* -t in c 1 o cr. os CO © CO t- Ph ^ 1 Cl-.rtHrt«(Mi- rH CM 1-1 C! rHr- & I 1 o m 00 R 4= ■« w co io 10 o a in O C c a oo o m cm co oo c: o o g 00 l> CD ao CO O W CO CO -O CO ■-* CT COCO Tt co -+ ir- c 1 "" t> CO I'HCOOCC CM 00 O O r- IC t^ rli C- CO CI s CO 00 t~ oo cd . ^ CO CO CM CD ur CM CO tJ c- CO CO CM CM C-1 r- CO CM CO CM CM CM 73 » « & ll °G CD ■W -M CCCO-Pf- TfOf c OS ~i< CO O CO CO c CO O CO HI o o o a r-i CD r=j £ Oi OCOOCs. o *f co oi 6 t> c^ O CO l> t^ o a IT CN cv CO t^ ~* OS ^r-occc OS Mi -f OO CI CO CM CO CI 00 l> os' OS ^wwco <- CO 7- cr a CM ^ CM CM CM r-1 CT CI >5 s U ts CZi O a so CO CD CO CO CO IT ic oo cr CC c i CO CM -* OO o -+ c CJ O CO CM c '3 £ -rf CD CO Oi J>- CC co -* oc a | t-CO ■^ oc io os F CO t— iQ rH CO -# u '5 •O OS CO 1(0 CC i— " c O OS ir cc (M CO its co cr. cs a CO d d o o o £ CI i-H r-1 rH i-i r-l CN rH C] rH CM CJ CM Ca rR o H H CM *» N *-< iO «-* t« e^ coca c -rd li- 00 OS CO OCC OS cr. co ~i> o o o S O t> CN ■— i CO rj *# o a- cr ce t-CC t- CO rH CI 00 oo >n t~ OS rH rd V CD lO UO UO CO 0" CO CO -^ CO CM CO *# -^ ^ oc 1 CM 00 -rji CO iri in co' < £ c- ■et CO -^ CC t^ -f OO CI CI CO HI o o o o t- 00 CO CM iH CM b- ^ *& t- lO rH 00 y o o o o o c c CC XOOOOIOHCC 00 i-H OS rH O* O OS ea > ft. O O O O — < r- ICOOO-f-f-f^ S7?i cs os cc. ~. ~. cr CD t^ t~ 1- 00 00 00 0C t^ oo oo oo x x r ooooooooooocoooc ■ 00 HHr- mi-^^^T-ii— c 5 « CO H S5 w H fe o o a 0 R H o f -J H ~! ' i-1 CI ' ■> >i b/j be bi) M b c« "3 5 5 s 5 = a J co a ■a^ p c a • CD J. CD CO a 1 " ca a D 5 3 g III ) o n u CD CD CD S CO CD CD R 5 m .2 9 = +- c p I 5 fl » i.a * "S M co .S c3 f> a o o CD 1 rn CD CD CD CD 1* G W +j +^ +- +J +J 4- o o o cd a c SS< P CD c8 2,FH C. cSCCto^-fOc Sr3j ;ja R bjo c I* r 1 "i cc d3 : d th cm co *h its cc f a riCMCO^lCSCOC-OC OS O i- 1 (M & CHEMISTRY OF COTTON. 123 co-** toc&t-oocsoHcico^iia ,-, _ h h h rt h :j w :i ci ci tj tH nco ■<* CO t- QO © O — t a b :tel CX< EH X 00 3 ' «a Sx'.r.~ = d x " x x O ^ 10 « : x >_ - ^ c S S5 «=--<_ cs =s"313 c © ?: aj a 3 x * If E - o 2 o 5 £l <&< fcS ■^2 a .* X x no J- £ pq £^5"~*xx « -^ tn -*-; — £ 3 am _ ax s= : a = =.= t,Ht]M = J.'«H«fi;o • ■ c 3 u 3 t^ « i-i a •-< co x co ? i ^) © C*» 00 — ** - l^fWCOffit- CM CQ *i C3 r-i CM X -f CO •— co o a t» n m w o tc t* ^ COi— IN iq co lOOOMOlOrt-^OlflHt-OO -+ © ©" rH no t~ o" c^ ira © — i x ! CO CO -* CO CO CO i— i O CI — '■— ©-*©«-* — ■ x -h o :i it. o cm x o r. x ~+©COCO©©'+©ClX© CI N CM CM i—i CM CM CM CI CI i— ©© DTflOCOHAOinCOOlQ X CI 'tlflXW-tl'Ht-rit-t* >—" co t> os io ci © ^1 co o ©* i-5 1*- N© 00 © t- CO CO © OJ C~ X X ci ci ci • © X CO x © • ' t-" x" C 'lO--+-*-*COCOCO-*CO © oocoi>r-oococooc t- X CO t^ CO © C: t- CI © L- © CM* Tf *5* CO -<# CO ■** CO -1* ■*$ 'CO* tjh © © © ci tr- n i~ © r-- co © -* oi rt CI ©-f^-CO©©CJ»CCICOt> BO CO CO ©* co* m" CM* CM* CM Tj" CO CO 00* -f ? i © x ■ ;o t- — t- © 'O r- x r: ©ioxio©tr-CM©©©x-* CJ T " C] C* ci i-" CM CM 1-" C3 C* Tj* © © i-H .-«- f-tco© ciocoiiaoio do r-i © ci oo © -t c-* © oo i> ci ci '-.ai.-. \\ * t~ X CO || r-» r-t © © © LO X LO -?■ C* CO © .O © © i— I tr- in 01 ~ -f L- © Ol CO © CI © ci © ©* x" i-* ©* id txO* ci i-i t> CQ w © © © © ?l CI © © © ^ © © © X X © © X © X © xxxxxxxxxxx x ci a CO Q B XX X S D tc - 9 0 ^ bC « P - bjj co a = Z 3^3 c :« rt"Cr3 co^d cs x c;rOrw oo If? >'3gS 5 zi f. a f*3 H o n n - ?K1 ?1 -f X O 3 x Ci =: ~. ~ r; * — — . r. x x x x xxxxxxxxxxxoo :i=> & iftCOt— X©©-JClCO-1'iO HHrtHrtNClWNClN Ot~X030HC-1M-ri 124 THE COTTON PLANT. OCOOHN P. A MM 02 GO * c3^ C«G0 O ad" T3gg ©00 2 a W P>§ «53 gt> A a erf — P p* HPg P< Pi Phi DO S) i1 •5oo +- g S ™* S o I o §^ goo $ * 5hftp5 hj3 .* ti ■ tS"3 -r- X — I p P P P J3 M frcM § p p p Si cs o « » .jh -3 .- 5 o •fc P-S'S'p. 3 3 - to pt> M£ Pg IS ■wt^lOQOH't ■* OS 00 CO t-CQb-C in oi ei r-i c4 oicit co t o p-3 iTJ CO* oo m oo i* !M OS m oc bJD® § i co" CN —5 ^ cdoi m ro i m m h m t* ;cs Ht-'* co > co m cm © (M t-h m co CO* CO C^ Tji -* co co co e. f: cs COCICi iH CO © CO CO o 5* O CO CO L— tJ< w in" in in tji co o co co in co c o* oo o t- CO 1-H MHO co -«* ci ■W in iH OS OS CS OS OS © N •** CO CS OS OS CO CO oo CO CO ~ s j5 W I g & a.g g a M o d a 6.1 6 A g A H <1H 00 CJS O iH j^ cS • .. cc cm" >-. -• (-•aoo-aj; a poo^lzi Eh ca &^ 3 rs -W S f" 3 £ J £ -g 4) « ^M | <£ s "§ J E'-So a02 f-1 ■ fcc ft CD — N C3 © © g BO CO 7! 2 "3 - >.% go 5=2 3 02 fc H •goooooooooooooooooooo-2^o--ooooooooocooo ^rd'w'dT!,w'C'3'w'5'd'0't'C'd'C'w'3r3'd't!^,c;'^i3'^r^'^'^r-'^r^^r^^,^r-'^ in c; 20.58 37.84 29.64 c o ©©C-t^COCOCO©CM OiniflCNIOHrtirjCMCSa NHOffiOOHC£!^«OHlOaCNO -tHMCocooot'CM -*oocoaiHp:i>ciofOiN osioccocoroocMoc^most-t-os t— CO CO CO CO CO cs © » « iri o* i> (O 50 « io co «d ^ «0 od c^ o irf t> in ^L^L^co"^cot>cot^c>:oco'Tj*rCurS i-h .i-^cdt-t^ g CM CM CM CM CM ■ CO Cfl CM iO g ! C o © CO 00 OJ T-t co o O O CO fficoo Trl CO OS OS CO OS t- OS co m CM CM OS © CO CO CO t- CO CD © i-H © CO t>TiH CO CO CO CO Tfl © ' © CO t- CO © OSlO t- CO lO '©©©CO c3 t- m co O CO OS CO © uO ^ —1 ■■* L- © rH CMTfHI>»00 0 CO Ci CO CO CM CO CO CM HO CM* i-h CO* t-" CM* CO CM* OS* CO "* CO CM i-H CO* CO "■# O CM CD* rH ^ UO t^ CO CO* CO CM OS* CO* CO CM CO* CO t?n ^JCMC^CMCMr^C^CQCMCMT^CMiMCMCNCMCMCMCMCOCM i-fCMWCMCMCMCMCMCOr-lr-i^-tCMC0i-t 2J" CM CO rH O O -* CO t* CM © "* OS "# OS CO cc OS i-l OS CM CO CM CO CO OS CO t- O O CO CM O O CM W LC CO CO CM CO lO CM CO CO OS CO CO ■* ooooosmt-c-^m©©t- s, CM © L0 i-h CO "* © t- W © iO a t- ■* in cot^cococoot-iococoinioco CO co ta to lO COuSCOCOOOlOirslOCMCMCOCO <* t- TH •H CM O th cm tji i a OHCJ r> co coco || cococotooo-^mcooioioo UOi^iO^OSCMCOiOOOiO^CMOS CO © © © CO m t-- m CO co©-^os©©©©o©uoo ©CMC0COC0lOir5tOCM©0QCM CO CO CO tDOWHHTtOlNOHHKli-l ■H © CM OS T* ©* v-H* © ©* i-I CO -rj* O* ©" CM -H ,-* v O 1Q t- t> t> t* t- OS »0 O) CQ CO -^ ^ 00 1ft CD t- CO Q0 00 t- t-t-COHHHHHCOOOCOaOCOCOOO inminininmmmcit>t>i>b-t>i>t-i>t-t>t>t-i> t> i> t- occococoocl— c» t- 1*- 1- i> t> OOOOOOOOOOCOOCCOCOCOaoCOCOOOOOOOOOCOOOOOOOCO COQOXCOOOX-XXX'XXCOWOOOO 111 THCqC0-^iriCOt>00OJ©-HCMCO-*lACOt-COOS©rHCM i'*m©t--coos©T-icMco'^in©t* CMCMCM CMCMC^CMCMCMeNCOCOCOCOCOCOCOCO 126 THE COTTON PLANT. OOOSOi-iC1«'tlO«Dt>OOC3CHiMeO** uo co h x c o h ci a £ is sb , ■"""-"ts P.P-1 **a ■p cS g £> M .a® pi c$ sil 03 o 3^. Ed .mi4 s 03 a -^ § 8,3 o +5W a00 § O P co ■»- O CD « 00 ^- N O h* O CO ifl N C © O QO O ' - C I * I S^Ln*fOWriC0030)ifiH00m(DHCDifl -*ooo O^^*^^^lOinc0^Tji0SC£JC0lOLCJC0C0" COCilO ;t>t*ot'h-oi>co©Hoorit'(> IHMOlHOCOC-CDOCDCOCXl- j a; uj w (j^ i" tt 3 CO CM CM CM CO t^ CO CO L-* tC irf W C) CD C. CO -+ r-l i-H CC O CO Ci co io co a t-- m CO :c rC CM CO CO CMOQ co'oiuo > CM ci m CO i-h CO O O CD CD CI CI CO O ~& CO : CC CC CI Ci ^ 00 -+ -* Ci Ci t- CD Ci CD CO CI ic4oQwdccoo6oJacd»fi^l**t-co NOIHWWHHHHHHHCXJNHN oc no c: i— ( co co -+ CO CO It- CO t> t^ t^ CO L-^ CD l> 't-cocico-fcomc-Nt-coiaooncooi ; CO C-J lO t> CC lO O r- lOOCiCO'^COCOCDiCCD CJ lO CO LCO -j5 CO: ci CD iri t-J ro* CO o co" CO CO CO L- .^CMCMCMCMCMCMC: ; o o -# -^ CO o co ci no co co lo r- CM t-i (M CM C] C) 0 4 CI )CO^(NOOCDOini>C]OON-*COCO HCCCJirHOtMCMCOt-COCiCOOCilO jidt^ioLOincduot^iocoiJOiocdco" *r lO O O O OOOOOOOJWNtOOOO CJCO'*miOCDt>t>HCDCi&CO irfHHOHOSCOCC'C CO CO coco CO o t- I CO U5 CM ^ ii^HOO^^ Nncd O >-h CI O 00 CI CO CI L0 O "4* CO CM r-iCOOCDCOOO UOiOCO HXOHOJOO t- "^ *-H A * CC'OOCOOOCOCOOOOOCOCOCOCOCO- t- c- l- i~~ t- r- i - t - x x x x x ' X X- 00 CO OO O0 CO CO CO DC X1 00 00 c O O r- o Ci Ci Ci C. Ci Ci Ci Ci ■X CC OC CO CO 00 as a eg a Sooooooooc«oSooc;o^-*o3cro-* 10 co t-ooO)OHn CHEMISTRY OF COTTON. 127 ;x^^ oifl-t oHr^o *f -f i> t? :: :i r: i- c i*. n r. x ^ i* ?i :i r; •* oo ^ oi 03 h 5 C. N O W O r; w t» i< M ed "^ > tjh t^ cd t- ci id x ci co •■* CM © CC OXJ)O^Cini> Olfttr-O(Mb-000i^*i © 't n ci ic o to a; co <«# b- ■<* »h w co eo t> CO -<# t- ^ co ci © © OS CO CO CO ©© © CJ O OS t+< ci co x x -*r ci x © — © o © H i- o o « M x t- IHCJMM7JMHMH t^HCOCOHMTf ION N s CO CM C] © O O 03 •■* ttWOHNH-iOHi*^© 'fHXO!Ot-l>f-f't^L- o o to Ci -* :i coo OJ (M M O OS ?1 -^ 1ft © CO CO CO © CO -# (M lO rH CO U0i CO CO .— 1 CM © -** CO m © CO CI CM © © -** 00 HoxachaxcoiOrt CO-rPCO-^COCOOSt^"^ Tjit^©-*"#idcioic6 *«* t- © © O O C- CO HNHOoooooiflr-HHinHOoxo3t*oaooaa OS © ~* iC2 © co uo co co cc H©Tfl>H i— ICNCOi-HC -f ci -*f © lQ o © CI C- lO OS r-* © CN t- t- !>• © CO CO -^ 00 ' J CO CO ) -* © LO ~+ CO lO rH ■— © O CI -^ CJ © © CO t-- CO © CI © CO CI lO H CM "* t— © LO CO -* OS fi COt^COt^t^OSt^t^t^L^t^I CO -* T— I C ■ CO tr^ ©* CO ©" l t— t> i— ' t^ >— i t-j c; i.*: x © — n © © x oi ro oi © ©■ © © co © © tp © i ■ CI OS o © ©CM©MO^i^Cl©^C3COCOrt'ClCOt>COi^l^-L^^in^COCOOO-^t-ClCMi-H I>c6aso©o5ooosc»oeo^o»c»i^odcocxji>t^osi^ ' CO CI ■ © ©" t-t^t-t>i>f-r— t^t-r-i^r iNO"i>xcoi>^©c:t-c:ccixxc. cct^r. aoooooHr--^ h hhh io i-O lo l_o i- i-. io .o i- i- lo i". t~ i- t- tr- t— t— t- t i^t-[>i>t>t-i>ccxcc x : X X X X XXX CO CO CO CO CO CO CO X X X X X X X X X X X CC X X X X X X XX X X X X X X X X X X X X X X CO CO CC CO c fan So ?* s « g ^ Cj P O oooooooooooocoo ooooooooooooooco Q H ; : ; ; : ;H ;C ;o g ;;;;;;;;;;;;;;; B §::::::::::;:::: : - ....... , - >: oq h=H C/2 ; ; *^cMcoM'00©©r^c3co^LO©t^co©©rHCi:o^o©r~co©©'^cMc^ Hr i h h -i h h h n h :i ckicj :i ;i m :i cm cJcocococococococococo^T^'^i-^-^fTf-^ 128 o THE COTTON PLANT. b-000>OHNCO-*iOCDt»COO>OHN«'*lO»NQOO)OH|Mm-*Jft(0 C0OONOO00OO0000^i/:XN«D^fll0005OOO0000N0000 g r-jt-osmoiccco-^coioco-* ■ win«Mccoocci lOOONiftmmHHHOi^noooj t- co t— O riO«)Nt-HOO(Dt-HNO(Mffi^O)aJW'#HmQOOO'* ooiiii t> "# m CO «iricaaw6d»flec>fci6NddpdMOMi>mt*cipcoH 'OOOffl ■ odojoc i t^co t> CD t> • oo to o no . lOCD CO IO CO * 00 00 t— OS t— pes rtHHHOffiCiaOOOOOOOOHHHMHHHCinfltlflNHHtKlNgCjncOMn ooooooooi^i^L-t-ooc^ooooocooooooooooooaococcoooooooooooooooooooococcooaocococooo ooooooxoooooooooooooooooooooooooooooooocoooooaooocoooooxooocoooocccocococccoco w I:::::::;;::::::::::::::;:::;::;:!::;:!.' ■< ::::::::■ •:::• '■:::■;::::::::::::::::: : : q I::::::::::::::::;:::::::::::::::::::::; ■? . t I:::::::::,::::::: ■ . ssfl g : : • ■ |§| i i ! ! i S^f! i i : i : i i i i : i i ■ : : i ■ i : ! i i ! i ■ ho^. ::::::::::::::::::::.,:::.. .:.... 0-i3 -P °a § ;::::::::::::::::::::::::::::::::::::: : c e p I ........ 3° is :::: b » I:;::::::::::::::::::::::: : : -u g :..:.:::..::::.::::.::.:..::. : o ■ • • ; ; « ; ; g o : : i : : : : : : i :::::::::::::::: : i : : i : i i : : : : : O §oooooooooocoooooooooooooooooooooooooooo 8 d .'•'■• •',* * '•'' ' ' ■ • ■ • ■ ■ • '•''!• • t*C0CftOHMC0^iOt0l*00CfiOHW«^O«Dl^C0OOH(Nrt'*ifle0l>000SOTH0:lC0Tj(»ft«Cl ^"^^lftlOlOi01C>ir3ir300lOCOO^t-t>tr-t-»l>tr-t>'t>OOOOCOQOOOCOOO CHEMISTRY OF COTTON. 129 !>QO»OHNM-fL'5COt*aiC3C^W^*tincl oo oo x os os a os i iCSOSOSCSOSOOOOCOOOOO- 'COO)OHWM-lOOI>COr- O ^ C"l CO ■ 1 eO fc- 00 OS O |H C too ot^x^osesrocont- — t< co co - <-noooooooooo O c "J © yj h -j; :i - o 'O r: ^ x i- cc c^ c m h co t- o o : oooooo ooooooooooooo cm coo in ci i-h cc x -j t- c r- r: -t lo :i c ^ o i r4 N CZ> t* CI M CC 00 01 CO -t ~* OS t- m CM - i -c o ^ ^ ;i m l^ -r c ■ 'C0C0)OCCCO(0CQHeOl OS c» co *■# oo oj oo oo in c*i co h^oxi co u-i d i> r4 d in d i- -p iri in x co in c ooo ooooooooo ■-* t- OS 00 C- rH t- on ia OO CO CO ■ I co co "*# cd co eo in -# *-" co in* c-i co co co ci ci c tH CD iH CO jin CD —I rH -* ci e O C: O O O : OS r-i X CO i o © os ro -i* m" c " ■»# -* •* ** «# -^ -* - :3J ^Tji^^^^Tti^-r^-rJ*'*-rti^-^^-Tji'rI«l_',Sin'^-^-^-^in 8.35 7.60 0.63 7. 05 7.90 6.85 7.55 0.95 6.60 0.10 o 00 S3 o so 7.85 6.65 9.25 7.85 9.40 8.35 9.60 6. 95 6.25 9.00 o - - od cc a c c c - at 'a r a - cc c a c? c cc s c C7 C a cc a t> c o c CN o c IT — 3 OO so cecOOOnnCOCOeOOOOHHHfflMMNCqQOOOOOOOOOOOOOHHHHHHHHHH^HHHH^'HHH CO CC CO 10 X X X X X X X X X X X X X X X I - I - X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X ooooooooooooooooooocccoocooocoococoooooooodooooooodo -^t — ^ *C — ^ — ^^'^-r-r'c-r — 'C^tt — ~r t ^ — -^ ^ — 'C'O'O'O'O'd'a -r ^r^rr — ^'r^'r-c'ccr^-; — .- — r*- oo os o i-h ci *o ~* »n co t- cc cs o «h ci co -f w *o t— oo cs o --< ci co -f m co t- x o o ^h ci co -* in co t-- x os o »h C3 W'tmci^co XOOGGCSCSOSOsOiOSOSOSCSO". O O OOOOO O 50HH^HHHHHHHN««NN«01NWNC(: EO»COC9COCOCQCa 1993— No. 33^-9 ^ 130 THE COTTON PLANT. aoHM«t it o ^ to » © co < '00«0«DM-l" » o ooooooooooooooooooo 00:000000000000 — c=> > oo cs m i— i -* os in t> ■•* i— i so m co" i> oo* *& *o co H* coj cq oi tfl eo c*s^ ^ ^ in tn co ^ t£ cd ^£ id oo 2>- • * a » A «8 >oo ooooooo o cr-i r> t- CO t>(M C- tJ* CJ ooooooooo O O O o o o o o 1 C p-i ."O X 00 t f CO lC CO C5 t- r-i CM : OS L— r-l woomtxi- o o t> co t- .n co m t im cc c; Ln co co -t t- i -* i.t n co -* -* in -f co co en DO co OOOOOOOO : o a o o o o o XXX X X x X X X X X X x ; t-mxcocot'^H eocDcotNCD^fcomia ^ t> i> w i> co © co" a a* d co o d c» a i> co x © c x oo w od < H h h H o cKi :i ;i ci ci :t T) :i :i : i :ki *i :i :i -:i ?j n fi cm w c X X X X X X X X /. X X X X X X X X X X / X X X X X X. X X X X X X X X X X X X X X X X X X X X X X X X. J. r, X X X X X X X X X X X X X ooooooocco ooooooo o o c o o csop-oico-tinct-xoo ' in C N X C C H 71 77 -t if! C N CO C! O H ( -+-*-+ -f -r i ::■ it »* it it m it it it it cct co c o co it co -o co cch.t-i^r-t>is- i— t— t- CHEMISTRY OF COTTON. 131 oj o ^ ?i ?? ■* to *o t- t ^ o h :i ?: -f lo o ^ oo c: o - ?i :: *f I*: c i- x ^ r h :i - -t l^ ot-QOQOHdcoHiiaeixioOiC i> co x m x x r, x x x x ~ o z r. r * a ^ ci r. c o o c ;■ o o c o c — h h h h p- n h r- h ci :i :i ;i :i :i n 01 :i ?i .-o i-l H H H H H rt H H n M H H H H H H H r- n H !M ^ ^ W CI C4 :i M :i M :) 71 M 7) -1 CI 71 CI 71 CI 71 71 71 71 CI 71 71 CM 71 71 71 OS rj* -* in to co -^ •■# - O O oooooooooooooooooooooooooooooo coo ooooooo oooooooooo ooooo ooocoooooooo o o o o o o o o •«# CD CD t~- iTD X •— ' t>- -f "^ Ot-^fOH^ffl^C) O 51 t Oi « ITD t— L- -rf OS 00 •— ! O CO CJ CJ I— C5 .— ' X CO i-h ifD lO W - iO ^ O "* t^ CD "# -»# ooooooooooooo ooooooocroooo o o — o o ©ooooooooooooo ooo 0)03-t»t-00rrOt>717D00O TjtOrJlfOiO-tr-dOinr- i .— i C- CM -* CO ~* X O CO CJ W O t ? -f r-( t^ CO 7: m CI CD-J-^H co*t>cBaot>t>co^aiwciocoxoococ»t^ 7171 CI 71 71 71 71 71 71 71 71 71 71 71 71 7 1 71 71 71 71 CI 71 71 71 CM 71 7 1 71 71 71 71 71 71 7 1 71 7 1 71 71 71 71 71 CD 71 71 7 1 71 7 17171 7' 71 CI X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X /. X X X X X X X X X x X X X X X X X X X X X 00 CO X X X X X X X X X X X X X X X X X CO X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Oi O T-" CI CD -P ut CO t-- X.. CD 77 — * 71 00 -+ U0 CO t^ X O) O -— ' CM CD -* HO CO h- X OS O r- 1 CI CD -* LO CO t- X CD O — 71 CD -*■ i.7 CO r- X O O - _^7777C777C7-- — •— ■ — • — ' • — ' — * — — 7i ?i ri :i ?i :i "i :i n n " 1 71 CI CI CI CI 71 CI CI CI CI CI CI CI CI CI CI 71 CM CI CI 71 71 71 71 71 CI CM 71 CM J(M ■^OOOOOOOOifi-fCC'CC 'O 71 oo oo cocoinoo S to t- to cc CO co c cc t* cc ci m h m H I in m to in i > tO Oi -f CO CO ■-( CO M< -^ CM CO m CM Ol rH ■* t|I » -f m co ci m NH i-H t ■lOQ-fOOOt " ci in co o co ci co co co CM -—I CM CI C I c'l Ci C-1 CO t- o to o to O -+ CI 00 co ci t- cm to Tf cj r-i t> co in co" to -* o* in" o* m" in co tjh co -*~ O O O O -* O O -f CO CO CI X' OO CO CM CO ••* ! ~COHrfHL^LOr-i-tH0C ;. m :: in c h oo c co ci co co h *» c ^^^•^f-tcOCO-^'t'^^Tjl-^' 'C5OC0HC000-1'00: J co -j" ■** l> t> t> 00 I CO in CO CO - o o o o »n co co i— i ci CO' t- ci ci t- o ; i> co m Oi -h m in co t-- '- m co o; t- co i co co i> t> ci t> oi o* ci ci oo l^ oi co co in ^h in ci cm i-lt-00 tOH CO" CO O C5 r-H t-I N C) tH CM r* o to r^ co rH CO' ci cio ci Ci to" ^ HOG -f< Ci CO o -* © *«* C) o ■<* -J"" iH O O ~f © l> C] to CO to" in co t- coco ci co* o o o o co" t^ -f o CI Ol Ol CM CO 'O OO -f -f -f ?0 -f lO C I- ^ CM CO -f -+ in to CO ■X CO CO X X X X X X X X X X X X X X X X X CO X CO CO CO CO CO CO 00 X1 OO X X X X X CO X X X CO CO 00 CO CO CO 0£ o o c o c <±> •j. i. a- J- u^ ^r — — = — ^. =^ 00 CD CD 0? CD DC k tc p; a CO -„ CD ple'cc -*£ § H ^ c - ci 2 Sflfl§pSc»iflc3c;dfl ^Ci C= ^ rt^^^r-J^JO ™ ^ lO CI n 1C rl i* "'1 i n ri ^ LO U '=/. u tf ci. i/. cX cc * bf c/. c.f_ t(.i . ci ci cz r-. :z ~ cz X r X ci x <$ ■ ©COCDCDCDCDCD'iScDCDcUCD £ — © ft CD CD ■ B-.S p 3 f « s "3 w co 2 © © * co © a P ■ g & 2 1 u c s g.g g its « h o) ro -^ i- to r- x co o h ci co -t i- c t- co ci o h ci co ■CO' :o co 'Co co :: :: r: :: t -t -T -r t -r t -t"t -f in in m uo (NfMCNiMff-KMCJCNC-JCI CJ CM CM CI CM CN CM CM CM CM CM •7-1 CM CHEMISTRY OF COTTON. 133 'JiflOb«OOOOHaM'tL'jat»000)OHN*-+L';--NXSO-HMr3^i.'),JNaiC50HKO^ ij h h 11 h n h h h h h ?j ?i 71 :i :i ri :i :i ;i :i :■: ^ .~: ^ r: co r: r: r. r. -^ -^ *t t ^ 3 -tcj-rf ro r- :i c c. c o t^ -^ l': m i- c - co oc o in cc — m t- c :i x - * n ?i x r^ c -t ,& O Ci cq H CD a 'J Ol C3 Cl ^ 3 L- 3) o c m -t a t* x :i OJ CM x .-o O O n L- m ^ -j X T 1-1 ■*-> © t-' o x :i x x o t* i> c x d oi t.o^ c c a in d ^ d ir: ^ =' r* m uo^i w - :i io -t l* in m" 2 HHHi-tr-tHHHHrtSn : lOcciOcococjcoxt-xr-LCX ::cco> oo >— • co o cc co co r-i 5 i-H -* co l- c-i i— co '— oc co co co c i -*■ os t>- os iH »n i# co -r in l— cq cc c ci cc — E> co co -+ o m x w co : j 1-1 go © r-i ro n co c-i o* o i-j c-i ci ci .-H © OH,t> PI NIOM>C1'-0-HXX^ 01 -f t- iO H Cs — I CO t- I- CO t- 01 Tl 01 W N -f C CO 1-f g«iom©«)NMBJN,*Nt>»oNO'«fiQOt-coiooHNi-riin3-tcxi-A.in'xooic:iC):] o ril io ^ »o co co 6 i> « io ** ci i!*. « d io -*' li -t i-i m 'f 6 d d «i '^' in c o" h* «b io p: d 6 ^ in d in (MOHXHrt CO CO co CO r ; co co m eo co ; IQOOIO CO pCOlAt-COHCOCQCQCACOi : r- x i.o -f ~. cs h -* % i* -j ^ o x -f ^ :: x q ^ :j x '-o i- co f ^i ci > -* iO -r -+ CM PI r- 1 tO 0 > M CO -f -f CO PI CO ' I CO -f CO C". X t-h ' iT*-ji^-^coco-t<'*coco-#~?<'*co^- i Q w i-t c c. :i o c o ■l CO' -f -f CO ■*# 01 01 ti 01 oi - 1 ?i oi n :i:i:i ro co co co co co ■ co ro ro co ro co co ro ro ro co co co ro ro ro ro co co ro co ro X X X X X X X X X /. X x X x x x /_ X x x X X X X X X X X X X /. X X X) CO CO CC GO CO 00 CO GO X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X XV XX rd ^ :::::::::: I I::::::::-'::::::::::::::::: : aj c 3 «::::;:::::::::;::!::::::::::;;::::::::: "5 o ::::: ::::::::::::'.:::::::::;:::::::::: : .- a I §1 jii!iiiiii-{i! jiiJiiJiiJiiiiiiijililMi < +? • s I N M 1 1 N M I N M M M M • N \ M M N M Ml Ms h §;;;;;;::;:::::::;:;:;;;::::;;::;:::;:;: z "w .':::;:;!:;!!:!!;::!!:::!:!:: :;:::: i '.'.'. : c o I;;;;;::;:;;::;;::;;;;;;;;;;;:;;:::;;; ; H rj ■ ' § ooooooooooooooooooooooooooooocoooooocooo 03 ; ; :!!: t ;:; i ;::'.!!.'.';;.'.'.':.'.':.' .' ; ;!.'■;.'.■ ; ; a :::::::::::::::::::::::::::::::::::::: : wcot»»csoHNW^LOo^coooHM«^LOO^coGOH:->rj-f»n»t'000}3HNn^ T|i^Tji-ttOOiOOLlL-JL':OL-A';CCOOOOCOOOt-t-t-l-[>t*t>l>M>XXXCCoQ CHEMISTRY OF COTTON. 135 iftOt^cOoSc^^ojco^ocD^x^cD^o^co-^; .'riNXDOHC]M-to'oi>ooooi-'C>^-*in'Ot^xr;o-'?ir:Tfoio ooooqoooqo". sr. r; ~. ir. roc-3 ooooooooohhhhhhhhhhmnmmi i 71 IN ?i M co co ro ro co co d o c o c o QOXOOOHt-.'XXLOONOO O O O t— O ■ : -f C t^ C 2 T O ■ d cd x x d; o i cd i- c- x co cd — lo lo -r cd cd x -r cd x x c~. x c- — r oc — o — ; co -* OWOOOOOOOOiOOW '-' r: c. X r. c :i — co co co ^ "*-^OOc:OTj«^COOr- ' Ol -# CD ~* C: CD < woooocoeowooi H lO ^- CO CO O -^ - 21. 51 22.13 17.20 23.89 22. 89 23.73 22.12 00 OS CO i> co oS CO (M o 21.50 21.40 21. 66 15.78 19. 04 16.45 21. 53 o ■ CI ■ 3.58 3.20 4.18 2 4.65 4.83 4.24 00 00 o W l> l/C- CO I> CO •* CD C CD — O O — O ~ — -t -+ ~f X C*l O t- — ' iO X CD O O 01 01 CO O lO — I HACOOOOOQCO oiomto^i- i " t- l^ CC O Q O- lC l- CO CI -f tj« I- t OHHOOTHcoaxoifi OQHOOXXOf 3 iO co oi -r m cr. oi ■— i co -+ ■ ) -* CO CD 01 -— I ~* I > CD CO CO tO CI CO t~- 01 0 1 0 1 CO -^ O -# 7.41 7.64 6.94 5.42 6.99 7.34 7.50 ■* ~ £r CM 7.12 6.90 7.02 7.10 7 (i. 55 5. 87 5.70 5.74 6. 90 CD CJ 9.63 9.63 10 8.61 8.87 8.87 9. 00 o o o in m t~ t-^ CO •-* o 11.24 9.50 9. 34 , 8.10 8.50 18. 52 8.80 8. 20 9.30 9.49 cococococococococococorococoo i^-j-4---^H-'^-^--H^-----i-iHr-r-^^oo-;'-ti 0 CM CO -* usco r» 6 corocO'HH'rrrr— r-—))-rH -X -* -1* -^ m m US in m in m in fc , • "u'S 'i o> '& ■•so" rajj" "CICnT m . ra r^ co" »<) >■•«!,>•.!» Jk.^ m-2 "a B O ftg -1 &*5 E-^a a 5 1'ti 4« co 0 £ 1 is 12 1 gii C 3 = a 1^ 4 -Ml t-< "3 :o B ^ a a a a 1*!S art a a a CD Ph 5 us £.0 a h Bui a 3 ?.« 0 • M »JS ft{? P-P1 ^ .^ .7'^ rf.^ .of p ft ^* £ C" rSfi^ft 3 c8 n^ go t3 of a? Tj .of 'a.ofTi of +j =c aifi as acg aca ac fljg cc s cst: rt £ ft«s s n a c a C3 3 ci2 w>3 w>3 ** X 2 W CO 43 T. la •2oooooc_oc PlCC 5 tC ■* 0 sm n 0 S!C s BIO a it . 5 a 5 CO ^ in 3 .S S .S 9 .2 Cb " O " CD " ft a. a dLs a p. : g.2 a a a a v CU *M CD ^^ H h> h> hi ^ i-s ;Ha l-S r. "*JcOt— t-©-f<~#C"l-ft^ o o cc rH us t^ 0 t~ CO 0 -* CS -f 0 us jpirtin'+f^-ciajcc -h cd in 1 © OS O CD n in 00 in cm r-t m in cS v io cm o'ioho'ho'cc co o -* 1 °S PO r-J rH ci ■* in ^ iH -1- CM CO , *Hi— 1 r-t i-t »-* r-i rH rH »— | CM rH rH " ' ' ' ■" ' rtrH 1-1 rtrH Hrl rH (5 1 a .1= ^ CC rH CO -. CM CO Ci CO CM t- 00 (M CM O OS © "S s CT -* OS o 1-1 10 00 l^ ^ 0 CO 00 CO CO rH ru -S1"^ c, OS* -^ rH 1 . d co ci CO O) OS ItiH 1-H -rji US V CM CM CM CM C) Cl M CM CM 1-1 rH CM CM Cl CM 3 fe a £ I CD 1-i a CC OC 1 ^o 00 00 10 CO -* CS CO OS OO CO cc 00 CC 1 © rfi 00 CO CO CM 0 rH US t~ CM CO a CD rO « us CM CC if! 1 CO co' CO -** ■* >* ■* 00 -t CO us ** s I? T3 rH -¥ OS IO CO CO CO 0 in O rH OS O O "3 5 00 O CO O CC CM -ti CO LC: O -^ -* to 0 in c- CO CO CO CM OS CD o e> © CO »-h CI iO -* CD "** CM 00^* cc CO ^ >o t-1 CO CO* CO G> o , -^^irj-^-^^^-rn-^i CO io rji -*-*-* -* -* ■^ •^ US ^T "^ -* ,C| ft 1? Hi Ph •*oifiiT-it— Tt oo o t> 1 lA c-^ t> CO* t^ E> 00 co* in co' 06 eo ^ £ *^comrHcCrHCMcocDc> cm © t- 1 ■* O OS c- r-O 0 CM 00 t- CM CD ~ r-l © l> CD tH H 00 -t O* © t» OC (N t-H CO 00 000 OS CO t- US CO ^cot^cioiosodooooQC © WCC II c~ 00 00 00 CO oj CO t-' us 1-^ os 06 c2 £ £ II - =i? . io t- cc cc oo oo oo oo t~- 0 0 0 Cl CM Cl CO CO oO'XCf. at at at oo x r CC CO 00 CO CO CO CO CO 00 00 00 S a'5 oo oo co oo oo at' at x a 00 00 CO 00 CO CO CO CO 00 CO 00 £ 5 E ; . a c 1 1 <4- 'd | 5 a « m "3 "^ t n c § a 0 ■H CD c cc .2 M a W O c c- p - c3 u « HH © Q 3 3 a & ~ a IB S c 'CO p ^ "S &f- '3 M a -. .3 ce t. H rr<, ~. H oo_ooo_o_oo"9 3S< I 000 0 0 0 OO CO 0 TH-rrJ, rTjr- r-J cd ; c o * (i £ 6 t-QOOlOHINCO'TfL'- to C- 00 OJ 0 1-1 CM co -* usco r- ft COCOCC-*'*-*^-^^ ■* •* -<* in us m in us us us us rH rH tH I- •" r- " r- r- rH r- CHEMISTRY OF COTTON. 137 uO. © © © © tD C C 'O B oo c .5 JJ o I'd ■ "5 *"S hi 13 h a ft U 00 H T~t ' a h; ft a« ci a a: i ft a oo" a ; - ce oo ca I OO s 00 i) I 00 -[1 1-1 -rt U a ft2 ©OO to _-^> ~- 3 ft ."1 CD ea ft S j isH~^ a1" «-; o 3 'gpq a a to © > irs co tj* CM CO OS CO 00 o oc to 00 CO o 00 00 00 t- 04 Oi CO CO CMt- CM o IM in OS CO rH X c- o o 3 CO o rHi-i CO i-i H (M CO Ol « rH r-1 i-l i-1 rH CM rH CM TH to 8 o 33 co oo T-f CM ■-1 H OO 00 00 CO CO CO i-H OS'* oo -* ITJ 00 OS OS CO 00 CM CO CO O CD t- "# ^ o Oi CO Oi o OI CM 00 t- to t- Ci Oi t- CO OS CO CO co co vn CO CO CO C] i-l CO O00l> t— 1 CO t> t- CD OS C3 00 c- in o CO C- 00 00 o co in Ci o -r CO CM CO CO OS CM CM © © -* ITJ -^1 oo m CO C- CD OS CM irj ^* ^Jt Oi co in o co ~ 00 CO rH O lO 1-1 (M HflJ CO CO CO lO o t- r-tCO^f) t> OS lO ITJ CM CM i-i m CO c- 00 00 Oi CM CM i-l i* C~ CM CO CM Oi CO CD CM CO t- rH rH CM 00 CO CO O CO (M (M CO OS CO CO -* CO CO co c-a "# CO CO -^i CO ^< 1* CO COO "I* If CO -># co ■^1 CO CO © CM CM O -<* CO CO t- O H COO OS CO >* O CO t- co oo ia, ■«* irs to TJ1 Tf CO Oi oi •* o Ci Oi 00 o CC CM 00 in CD CM 00 CO © b- [- CO rH CO CO tr- tr- CD tr- CO CD CO CD CD o CO c- 00 CO t- CO c- CO t^ CO CD t> m IO CO t> CO CS iO CO t-l> ^ o IT- CO 00 t- 00 c-co CO Ci CD CD CO 00 kO 00 QQ CO CO CO CO CO CO CO 00 CO CO 00 CO co © tr- t- oo cc cc CO CC CO ) CO CC CC CO CO — a ■a rn'S^'a :^o .2 S g 1m< ua co © © © © © © © I rH HHHHt O© t-00 138 THE COTTON PLANT. CO ^ lO © r- CO O Ci H (M CO CO CO CO CO CO X O) O C- O CI CS Oi CS OS OS o c mtOt^COOJOH (M CO -i< JO to t- CO Ol O t > o o o — O O ' > -* 10 co r— co os o CI CI CI CJ CI CI CI CJ. CI CI CI C J CI C« CI CI CI CI CM OOOOOOOO-SmOh 5 ggg, IP, S^H i-°ri (- 3 _ p,°° « 'SI'sS'pVS §.2w a^ 2 Jfc- j fi.ej09 S 2 5 ii m d n S ■ oo^£cti,T3'-lc;00a;ooa)OCooa;oooooo DQ Om -S O ^ - -t io cc. ci CO IO N W CO -t oi o h CO CI 1 r-t CO CO ~* CD CN CO CO CD CO lO CJ (M (M CI H L- CO O if) rH t> ' ci o ^ ci cs co* i> ac5 ci ad o> w5 os* <-" os r^ oi oa aS os ■<# ■*# ■*# rA » > os io as o co cc x c t ># o i~ o -f a ^ o co co co co c ; o ^ t> «o ci yj i- ao -t o r ■ ! M CO H -J< ift X Cl H ■ > io co cs os c i c i c i ci as co * C l !N ri H M (M C^l H CN ! CM CJ r-f CI CI CI C H CO L— OS — < Xi t- tr- -rp i— | 1-. co to -r> i^ m" as" co os" in o* -J* -** t> cd -* m* in co ci CM CI CI CI CI C-l CI CI CI CI CI CI CI C) CI CJ .-h CI CO SlM'tOH^t^COiCacOt'OO W id O CO CO CO* CN CM CO CO CO* CM -* ~ X — O -h OS lO r-i -+ CO CO ! CO CO CO I* -f X lO 3 I- CO cot- CO CO -^ X CD ) f lO CO O CD CO CJ OS CO CM O ICO t> CO CO <0 CM ■ CO* CO" --l" CO* OS ^J" lO CO -*t O CD "* CN t-H C J CO* CO ILOCOr-HCOlOCDCOCDiOCD j ci in i - m as m © ci o i— i t-h re o o ci x h co o ci ** cd co x -t co co in rc cs ~r. - coHin hcoi> occoi— i»ni— ic-iinmco-pinost— '^ii-hosc cp co t> cd as r- -i tr— cidc'd^Hriix^Din^ci^^ps^Nin -* -tf Tji -tf Tfl CO -HH t-^-^'^co^iO'^-^^t'^-rfTrTji-^'-^ ■ i-i t~- C-l LO CI CO I O C-l i-l CO L- CO rH CS . ! CI O iO © U0 ^ i i t> co" in in" t-" cd i> co in" t>" cd" in r-i C- CO CD CD rH t* CO t> in CD CD* t> CD ) O CI ci cs to c ■ o co ci cd in c > t> E» t> l> i> CO* t > IO *=* t- CS (M OS CO CO CD t- ?in'cOOC0 1>"cDt>t'"t--"r-*j ' o co m co m ci c - ci c; ci '-? Tf cs i >Cil>ioeoQOi>t>ooo-ooe t>CDCDI>l>HX rHOXXXOL-XXXOSOSL-l>CDI>CDCin o ^> 01 5 Q o so r— t~ t O0OOO0O0COa0CC>QO( t- O O O »H O 00 O O O O (M Ci ~. C. ~.- 00 CO ^c;«^ccn«ixcoco :■ oo oo oo oo x co oo x x oo • -£ 3 03 C -M CO +3 ^H [^ o o « a o c w- a) OOCCQOOOOC^O ^ ::::;:: : l^ .-3'tlC5CDt*«C10Hnnrli co oo co co oo oc oo ~ — c: — . ^ 03 w ^ CCOOiiOa ooooooooooo OS CS CS OS OS o O j -f in cd t- x ; jcoooo; ) -* to CO t- CO OS o i i— ( rH i— I i— I CM CI CI CI CI CI CI CI CI CI CI CI CI CI CI CJ CM CI CHEMISTRY OF COTTON. 139 ^ocac ««-fLOCt-M-Cr- ?] — m o t- oo c ca -* ifO o t— 00 Cft O l- ci oo -* in •- S • r-i 00 tj) " o >< : =v : _ i- - — " • ~1 ■ ■» iflMF^ ""• . to EC 3 ■ 2 :__ _H 1 00 ^-1 M P ! a ■ X XC r« w d Xfl d :j d x ^^7." 1.8 8 i. Bill. 16.. il. 87 it. 1890, p. it. 1893, p. •o2 :"='-^ « - • oo 43 oc oa _ x p rH -- '•a. : o •XI Stat Stat Stat Stat 7J ggSjp ' p3Dd 1 d d Sta. Sta Sta Sta ^j cS S : t. r8 RSv - — . *3 -t_ -*J ^_ jg • - CO . ■ . p p ohusel act cor chusel sota S c3 ci - 7. > 1 X 0 '. i CO "3 . pi. i - £ ', ? ? ~- "t - J; * 'So Carol Carol Carol Carol oo>c : : :(it,= s : : 2 :r : ocsmm; a o3 £3 iOC(SOOOOO ^ a it o m - a -t c a:ici:i X t— « c I re : Jt^Cr-X- -;ct>'# cai.-: x oo r. i -t o a a r- ( (M n ci ci ci ci ci ci ci n ci —i c; ri -i ci *i ?i ci ?i ci r-i ci ci ?i — re ci n re ci ci ci ci ci c c .r; m a a a t- x - c ^ o cj ei ri ci 7i re ci ci ri re cm cc co o io o r; a w c') o -t in a - x x :i c - c^ ^ - c. o « :i c n -- rt m t^ a tn ccco t^ x in t-i 't -^ Tf t- o -h x o i-e a x -* o ."0 r-i s in t- a .-o q :j ."C ri m -^ cocacocdcoeoc^TjicoiAcomiooopio co o o" ^" m" m' c a i> a m' io lo x t^ i> a t^ o ci o in m* • tr- lOHKHHWrtNOt- id xt^.-ocor-iciaojoco 1 1> t- m' tt tjI in mi t> iri l": aJ c a n o o o x r. r. c x i- x :i x h t- m h t — *■ ci nr. t- ci x in :i :i x :i c cit — f xxxt^t-HMicMt lOm-rorecemci * ci o* -r* i> a e^ -t oi r: -t >f -ji -+' ?i in d ----- : l«HlOnOeOHOHW«l ^ X Ci h' r-i C H -X Ot-tH-t- o re CO HH \n c. a m a x co i i-4C4 c4 iMi-h ci a;" l- oa i I -* CO "# ^J< "*! CO * OS 00 O O O O ' ■ t* co « x o i.- a x c :i r- r- a x ei x a: 7\ x o o 1-1 ; o o m> a [^ m c ci co x w co q co ^ co .-o a ci a -i< in x co lo o x o r] o r-i c c a • a- a> a d ai aV a> a a io^ d d ^ ^ oV i> o\^ a i> t^ ue^n a ^ o n a a a o in ; i iH X OS Cl CO X' Ifl ■ xacriocomaxaa^a t> ed ao" t^ t^ t> t> *o co cd cd o co" ClOOOO oaOlftn^MO ; i> co r- x c. co a r- a a m :i co rt [- co t-i o in Cl t- CO OS -f CO *-* CO r-H CO U0 CO t-C-HX10WO>LOOOOOt>t>t^i>c^^cdc7Jooint^ o co ^ o o c co s :: r-tifl n «t ■i" O O CD H ?1 N OS C t* .-0 CJ t- o6 oo od r^ "i>oit^cdi>odK>c> co ce cs os as o. ro — oa ci ci ci ei ci ei re co co co ~¥ cs as x x x x x en x ■ x x x os os os ■ • o: os oo os oo o: ~ os oo o. oo oo oo oo oo oo os qq x -co o. oo o. os oo oo os os os os X X X X X X X X X • X X X X X X • -COXXXXXXXXXXXXXXXX XX X X X X X X X X X X X X i .j o co co -- ci cm ci en ». a c r. a c. oo os o. os os oococccoooococoo rajr-r^r^^-.^- — — rrj^FTj^rfjirj^ffjrrj r— rotr-oo ea- ooooooooooooooooo ; nd ro ro 'd 'a; rs to rz ea eo -O' hw co-tin a n x a a h ci co -t i wcicici :i:ici:icicojco - ■xoiohn: ■ x o- ao —i ci co -f in tp -f ~r -^ -f -j* -t- -t -* -p i.e in i". >o i.o o m m X OO O — ' 71 j a a a a a a a a a a i- C* CMCM Cl Cl 71 71 71 71 71 71 C4 Cl 71 Cl Cl 71 71 71 71 71 Tl 71 71 71 71 71 71 Cl Cl 71 71 Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl 140 THE COTTON PLANT. i ©q co •*& m s© t- ac c» © rH ci co -* m co i— r - i — t— l— i— t I- 1^ OO X X CO X CI C4 CI CM CI CI CI CI CI CI CI CI CI CI CI CI CO s c o +-< P^^^^r^^r-rT- - p- g ^ ^ -g. „ or ■ P ■ • • - • i s r, s • -hi* £Ph « o m co c cc r^ o t — h co co co ci co :oomHr- co t> CJ l' c; t H ffi Oi CO r--' CO t» rn O) d CO H Cl i-i CD -* ci o co" ■*o -* co co ci co o o i-i i - co f - o o : ; O -~i Oi O CO II > os oi t> in (O i QClCOCM-^HCCHin CD rH id id id -I- --' CI ~+ CO CI CI CJ CI CJ CI C I CJ C 1 c: CI CI CJ CI CJ C4 i— i to co co cd ci ^^CDCOOCDt-h^OOCJCOCO t> id -^ ci CD ■<# id ■** CO CD t^ CO j m <— as co c: o o «x <— c- o — ■ i— co cc < CO CO, CO c s ci o* co co" c co-rjicoTp-^'^'^'^'^'^-^-^ini'rrco'^ JtOOCOOiOCOlOCOQOt ioCDCDCiidcD^t^O"^fC. lO > CD CD CI CD O lO CO rH CD CO CO - HO t* ~H ■*- CO m -* ~* ci co CM t-h cd m O CO CO t- n i>- L- CD CD CD CD t- tr- CD t- CD fc- t- co ^^ OI>Ot^O ^ CO CO CM t- •— CO CO lO t- CO m m t— -*# CO t* *<# T-i or, ci 71 ci »n in m' co co -5 c8.S jjjCOOOOIOOO C C. CO CO CO ~. CO X CO J. X ~- x x x cr. -*© o "*H x' 00 X CO X X X X X 0J • 1o o - o n - o o n O o o o o o rS^ *-« -: nr r: T3^< 0 S h m cc -t io co i> CO' * ~ ■— ' ci ro -f m cd • t- t- t- t- i X X X X i cj ci cm ci ci CJ CJ cj cj ci cj a ci ca ci ssh CHEMISTRY OF COTTON. 141 BIBLIOGRAPHY. The following publications have been consulted in the preparation of the preceding article: Journals and reports. Agl. Sci., vols. 1-7, 1887-1893. Amer. Chem., n. ser., vols. 1-6, 1870-1876. Amer. Chem. Jour., vols. 1-13, 1880-1891. Amer. Jour. Pharm., 1869-1871, 1877-1891. Auu. Agron., P. P. Deherain3 vols. 1-18, 1875-1892. Aim. Clieui. uud Pkarm., vols. 125-175, 1863-1875. Ann. Chim. etPhys., 5th ser., vols. 7-30, 1876-1883; 6th ser., vols. 1-27, 1883-1892. Ann. Landw., A. von Lengerke, 1843-1870. Auu. Sci. Agron., 1884-1886, 1889-1892. Ann. Soc. Agr. France, 1869-1870, 1872-1880. Ber. deut. chem. Ges., 1876-1892. Biedermann's Central-Blatt, vols. 1-24, 1872-1890. Boston Jour. Chem., 1867-1877. Bui. Soc. Chim. Paris, 1867-1892. Chem. Ackersmann, 1861-1873. Chem.Gaz., 1842-1844, 1852-1859. Chem. News (Amer. ed.), 1867-1869. Chem. News (Eng. ed.), vols. 24-67, 1871-1893. Chem. Ztg., 1886. Compt. Rend., vols. 60-104. 1865-1887. Consular Reports, 1880-1893. Jahresber. Agr. Chem., 1858-1891. Jahresher. Fortschr. Chem., 1877-1887. Jour. Agr. Prat., 1855, 1858-1860, 1870-1893. Jour. Appl. Chem., vols. 4-6, 1869-1871. Jour. Landw., 1853-1892. Jour. Pharm. et Chim., 4th ser., vols. 5-30 ; 5th ser., vols. 1-26. Jour. Roy. Agl. Soc. England, 1840-1888. Jour, prakt. Chem., 1867-1869; vols. 1-46, 1870-1892. Jour. Agl. Soc. India, 1838-1850. Jour. Soc. Chem. Ind., vols. 1-12, 1882-1893. Landw. Jahrh., 1874-1891. Landw. Vers. Stat., vols. 1-42, 1859-1893. Liebig'sAuu., 1832-1882. Proc. Amer. Chem. Soc, vols. 1, 2 ; 1878, 1879. Report Cotton Department of India, 1868-69* Rpts. Ga. Dept. Agr., 1873-1892. Rpts. Ky. Dept. Agr. Rpts. La. Dept. Agr. Rpts. Va. Dept. Agr. Rpts. U. S. Patent Office, 1851-1861. Rpts. U. S. Department of Agriculture, 1862-1892. Rural Carolinian, vols. 1-5, 1869-1874. Rural Alabamian, vols. 1 and 2, 1872 and 1873. Southern Farm and Home, 1869-1873. Southern Cultivator, vols. 1-9, 26-37, 45, 48, and 49. Southern Planter and Farmer, 1868-1881 and 1883-1893. The Analyst, vols. 1-18, 1876-1893. 142 THE COTTON PLANT. The Chemist, vols. 1-5, 1854-1858. Trans. Highland Agrl. Soc. Scotland, 4th ser., vols. 1-20. Ztschr. analyt. Cheni., 1862-1883. Works. A Practical Treatise on Animal and Vegetable Fats and Oils, W. T. Brannt, 1888. Aschen-Analysen, E. Wolff, 1880. Cotton Planter's Manual, J. A. Turner, 1857. Cotton Culture, J. B. Lyman, 18G8. Cotton Culture in the South, Loring and Atkinson. Cotton Cultivator, Joseph Gibhs, 1862. Cotton Manufacture, Andrew Ure, 1861. Commercial Organic Analysis, A. H. Allen, 1886. Die Techuologio der Fette und Oele der PHanzen- und Thierreiche, Carl Schaedler, 1883. How Crops Grow, S. W. Johnson. Oils, Fats, Waxes, and their Manufactured Products, Alder Wright. Structure of the Cotton Fibre, F. H. Bowman, 1881. Zusammensetzung und Verdaulicbkeit der Futtermittel, T. Dietrich and J. Konig, 2 vols., 1891. Experiment station reports and bulletins. [Last publication examined.] < Alabama College Sta. Bui. 64. Arizona Sta. Bui. 11. Arkansas Sta. Rpt. 1894 and Bui. 32. California Sta. Bui. 106. Colorado Sta. Rpt. 1893 and Bui. 28. Connecticut State Sta. Rpt. 1893 and Bui. 119. Connecticut Storrs Sta. Rpt. 1893 and Bui. 13. Delaware Sta. Rpt. 1893 and Bui. 24. Florida Sta. Bui. 24. Georgia Sta. Rpt. 1894 and Bui. 27. Idaho Sta. Rpt. 1894 and Bui. 9. Illinois Sta. Rpt. 1894 and Bui. 37. Indiana Sta. Rpt. 1893 and Bui. 53. Iowa Sta. Bui. 27. Kansas Sta. Rpt. 1893 and Bui. 47. Kentucky Sta. Rpt. 1893 and Bui. 53. Louisiana Stas. Rpt. 1893 and Bui. 32. * Maine Sta. Rpt. 1893 and Bui. 15. Maryland Sta. Rpt. 1893 and Bui. 30. Massachusetts State Sta. Rpt. 1893 and Bui. 56. Massachusetts Hatch Sta. Rpt. 1895 and Bui. 27. Michigan Sta. Bui. 117. Minnesota Sta. Bui. 37. Mississippi Sta. Rpt. 1893 and Bui. 31. Missouri Sta. Bui. 27. Montana Sta. Rpt. 1894 and Bui. 1. Nebraska Sta. Rpt. 1894 and Bui. 41. Nevada Sta. Rpt. 1893 and Bui. 21. New Hampshire Sta. Rpt. 1893 and Bui. 23. New Jersey Stas. Rpt. 1893 and Bui. 98. New Mexico Sta. Bui. 15. New York State Sta. Rpt. 1893 and Bui. 81. New York Cornell Sta. Bui. 83. North Carolina Sta. Rpt. 1893 and Bui. 112. North Dakota Sta. Bui. 15. Ohio Sta. Rpt. 1893 and Bui. 54. Oklahoma Sta. Bui. 16. Oregon Sta. Bui. 33. Pennsylvania Sta. Rpt. 1893 and Bui. 28. Rhode Island Sta. Rpt. 1893 and Bui. 30. South Carolina Sta. Rpt. 1894 and Bui. 18. South Dakota Sta. Rpt. 1893 and Bui. 40. Tennessee Sta. Rpt. 1894. Texas Sta. Rpt. 1893 and Bui. 32. Utah Sta. Bui. 37. Vermont Sta. Rpt. 1893 and Bui. 42. Virginia Sta. Rpt. 1893 and Bui. 35. Washington Sta. Bui. 8. West Virginia Sta. Bui. 38. Wisconsin Sta. Rpt. 1893 and Bui. 42. Wyoming Sta. Rpt. 1893 and Bui. 21. CLIMATOLOGY AND SOILS. By Milton Whitney, Chief of Division of Soils, U. S. Department of Agriculture. INTRODUCTION. Cotton production in the United States is limited by climatic condi- tions to that portion of the country south of latitude 37°. A few small areas are cultivated north of this on account of local peculiarities of the climate or market conditions, but for the most part the cultivation does not extend north of the northern boundary of North Carolina. Just after the war, when cotton was bringing a high price, the cultiva- tion was extended into many northern localities, where the plant matures ouly in favorable seasons, but the cultivation of cotton in these northern areas has now been generally abandoned for more certain and, on the whole, more profitable crops for those localities. Prior to 1860 cotton was cultivated ou nearly every plantation in the South, but it was principally grown upon the deep, fertile loam soils, found by experience to be best adapted to it. At present, however, cotton is produced upon nearly all varieties of soils within the region in which the climatic conditions are favorable to its growth, and arti- ficial fertilizers are depended upon to increase the yield or hasten the ripening of the crop on soils which are not naturally adapted to it. With the present agricultural depression, affecting the cotton planter as it does the producer of all other staple agricultural crops, there must be a contraction of the acreage if for no other reason than to give up the cultivation of the crop on soils not adapted to it and turn them to their legitimate use in the production of other crops better adapted to them. clima.te. In treating of the climatic conditions favorable to the production of cotton it is apparent by glancing at a map showing the cotton acreage that a line crosses the country a little below latitude 37°, south of which the climatic conditions are generally favorable to the production of cotton, and north of which they are unfavorable on account of the short season and the relatively low mean temperature. Broadly speaking,1 the mean temperature of the year is about 15° higher in South Carolina, Georgia, Alabama, and Mississippi than in Massachusetts, New York, and Pennsylvania. During the winter the 1 South Carolina Sta. Bui. No. 1, n. ser. 143 144 THE COTTON PLANT. mean temperature is about 20° warmer at the South, and during the summer months about 10°. This of course means a longer and a warmer season. The daily range in temperature is nearly the same in both sections, but is slightly less in the South during the summer, giv- ing more uniform conditions of growth, and is somewhat greater at the South during October and November, the ripening period of the cotton crop, which is an important faetor in cotton production. The daily range in temperature increases, however, in both sections as the dis- tance from the coast increases. The mean annual rainfall for the northern section of the United States is about 40 inches, while at the South that amount is exceeded by some 10 or 17 inches. The rainfall in both sections generally increases from the winter, reaching a maximum about the middle of summer, the autumn being the driest period. This larger rainfall and higher temperature at the South give considerably more moisture in the air. The temperature of the dew-point at the South is 10° or 12° higher than in the Northern States, and a given volume of air con- tains nearly twice as much moisture. But as the amount of moisture which the air is capable of holding increases with the temperature, the per cent of the saturating quantity or the relative humidity is about the same at the South as in the Northern States mentioned. The relative humidity varies somewhat throughout the year, but it is slightly greater during the summer at the South than at the North. The following are the essential features of a climate adapted to the cultivation of cotton: The season must be sufficiently long for the crop to mature. One of the most important factors, therefore, is the prob- able date of the last killing frost in the spring and of the earliest frost in the autumn, for cotton has a very long growing period. Cotton picking is often extended far into the winter, but the first killing frost of autumn checks the active growth of the plant, and the blossoms or young bolls starting at this time will not develop into mature fruit. The crop requires six or seven months of favorable growing weather for its development. The next most important consideration is the amount and distribution of the heat and rainfall. Cotton is a plant which thrives in a very warm or even hot atmosphere, provided the atmosphere is moist and the transpiration is not so excessive as to overtax the powers of the plant. The temperature should be high and the daily range uniform during the early growing period of the plant. The mean daily temperature normally increases from the time the seed is put in until about the first of August, after which it as rapidly falls, making two distinct periods in the life of the plant. During the first period of high and increasing temperature the plant should be in full vegetative growth. Any great and sudden range in temperature, or any prolonged cold spell, is liable to check the vegetative growth of the plant and tend to ripen it, which is very undesirable during this stage of development. By the first or CLIMATOLOGY AND SOILS. 145 middle of August the plant sliould have attained its full vegetative growth — that is, it should have stored up all of the food material it needs. From this time on a decreasing temperature and a greater range of temperature between day and night are favorable to the pro- duction of a maximum crop, for this checks the vegetative growth and induces the plant to convert the food material it has accumulated into fruit. The soil also should be drier during this second period. As a rule the rainfall normally increases in the South from the spring to the middle of summer, when it decreases, ancfthe climate during the autumn is usually remarkably dry and bracing. These conditions are favorable to cotton production. During the earlier period the rain should fall in frequent showers rather than in heavy storms, and the very best seasons are when these showers occur at night, giving with a large and well-distributed rainfall a large amount of sunshine. These are the principal climatic conditions favorable to cotton pro- duction. They are gathered more from the experience of farmers than from the meteorologist, for the reason that the relations of climate to crop production are so involved and complex that it is impossible at present to give any adequate interpretation to the meteorological data and to show their full bearing upon crop production.1 We are not at present able to interpret the temperature records and determine the exact character of a season from them, for while the monthly temperature may on the whole appear to have been extremely favorable, yet the temperature on one single day during the month or during one hour might have been so unfavorable as to have affected the crop to such an extent as to have a marked effect upon the subsequent yield, no matter how favorable subsequent con- ditions may have been. Continuous records by self-recording instru- ments will give much fuller information in regard to climatic condi- tions, but it will be long before we are able to give a satisfactory interpretation to these records as they affect the health, vigor, and production of a given crop. We must not only consider the mean temperature or rainfall for the season or for the month, but the mean temperature and rainfall of each day and of each hour, for these may have a controlling influence upon crop production. This, of course, makes the relations exceedingly complex. The following summary of tables, prepared by Prof. P. H. Mell and published in Bulletin 8 of the Weather Bureau of this Department, gives the principal features of the climatic conditions of different sec- tions of the cotton belt during the growing season of the crop. Table 1 gives the earliest and latest dates of the last killing frosts in the spring from 1871 to 1891, both inclusive, and gives the average date 'When the conditions are such that there is a tendency to excessive growth of "weed" a dry hot May and June to check overgrowth of the plant may be found beneficial. 1993— No. 33 10 146 THE COTTON PLANT. at which they have occurred. The heavy frosts at the South have generally ended by the 15th of April, and it is customary to plant cot- ton from April 1 to May 10, as there is then little chance of the crop being injured by the time it has germinated and appeared above the surface. It is not considered advisable, however, to begin planting before April 15 on account of the cool nights during this month, which are liable to reduce the vitality of the plants. Tahle 1. — Dates of last IciUing frosts in the cotton belt, exhibiting early and backward springs, from 1871 to 1891, inclusive. Station. ALONQ THE NORTHERN LIMIT. Atlanta, Ga Charlotte, N. C Chattanooga, Tenn . El Paso, Tex Fort Davis, Tex Fort Elliott, Tex... Fort Smith, Ark . . . Knoxville, Tenn.... Little Pock, Ark . . . Memphis, Tenn Nashville, Tenn THROIGH THE MIDDLE PORTION. Auburn, Ala . Augusta, Ga . Earliest. Feb. Mar. Jan. Mar. Feb. Mar. Mar. Mar. Feb. Feb. Feb. 2, 1882 10, 1884 25, 1880 7, 1885 25. 1888 2, 1800 9, 1884 17, 1800 22, 1882 25. 1889 2, 1882 Mar. 11, : Feb. " ,1889 ,1882 Charleston,^ S. C i Jan. 4,1882 , 1882 ,1878 1882 ,1889 1887 ,1874 Hatteras, N. C. Kitty Hawk, N. C. Montgomery, Ala . Palestine, Tex Shreveport, La Vicksburg, Miss . . Wilmington, N, C . ALONG THE SOUTHERN LIMIT. Jan. 4, : Jan. 18,: Feb. 2,: Feb. 24, : Jan. 12, Jan. 16,: Jan. 23, : Brownsville, Tex Dec. Cedar Keys, Fla Dec. Galveston, Tex ! Dec. Indianola, Tex Nov. Jacksonville, Fla Dec Mobile, Ala . New Orleans, La Pensacola, Fla Rio Grande City, Tex . San Antonio, Tex Savannah, Ga Dec. Nov. Dec. Dec. Dec. Jan. 26, 1880 7, 1887 26, 1880 30, 1878 18, 1880 27, 1880 26, 1882 27, 1880 16, 1882 7, 1878 4, 1884 Latest. Apr. 8,1880 May 3,1879 Apr. 8, 1886 Apr. 2'J, 1882 Apr. 22, 1884 Apr. 30, 1880 Apr. 5, 1887 Apr. 25, 1883 Apr. 14, 1881 Apr. 16, 1882 May 1, 1886 Apr. 6, 1886 Apr. 14, 1885 'Apr. 2, 1881 Apr. 5, 1881 Apr. 19, 1875 Apr. 6, 1886 Apr. 7, 1886 Apr. 7, 1886 Apr. 6, 1886 Apr. 20, 1890 Mar. 1, 1890 Mar. 12, 1888 Mar. 1, 1890 Apr. 14, 1881 Mar. 23, 1883 Apr. 6, 1886 Mar. 14, 1886 -Mar. 23, 1881 Mar. 2, 1890 Apr. 14, 1881 Apr. 13, 1885 Average. Mar. 17 Mar. 30 Mar. 18 Mar. 31 Apr. 1 Apr. 6 Mar. 22 Apr. 7 Mar. 21 Mar. 31 Mar. 28 Mar. 23 Mar. 18 Feb. 25 Feb. 23 Mar. 12 Mar. 8 Mar. 17 Feb. 24 Feb. 27 Mar. 14 Jan. 29 Jan. 22 Jan. 27 Feb. 7 Feb. 3 Feb. 21 Jan. 13 Feb. 24 Jan. 24 Feb. 11 Feb. 26 1Alsoinl887. *Also in 1885. Table 2 gives the date of the first killing frost of the autumn in the cotton belt from 1832 to 1890, inclusive. The detailed information for the separate localities is not published in the bulletin from which the table is taken. It is stated in a general wa^that at Charlotte, Chattanooga, and Nashville frosts may be expected as early as October 15; at Atlanta, Starksville, Vicksburg, and Palestine killing frosts usually occur as early as November 1, while along the coast of Georgia and Alabama they may be expected any time after November 15. CLIMATOLOGY AND SOILS. Table 2. — Dates of the first killing frosts in the cotton belt. 147 Year. Oct. Nov. Dec. Year. Oct, Nov. Dec. 1832 20 21 20 1864 1833 1865 20 1834 1866 1835 12 1867 26 24 21 18 29 15 7 1836 1S08 1837 26 1869 1838 1870 1839 1871 1840 14 1872 " 1841 25 25 1873 1S12 1874 1 7 1843 1875 1 1 1844 14 1876 26 19 20 21 13 23 20 1845 12 19 1877 1840 30 1878 1 1847 1 . 1879 1 1848 26 17 6 27 1880 1 1849 1881 1850 1882 1851 1883 12 1852 1884 24 1 13 16 18 20 17 1853 14 1885 1854 1886 1855 25 8 1887 1850 1888 1857 1889 1858 7 1890 1859 1891 7 25 15 9 I860. 13 1892 1861 ... 24 1893 1862 1894 1863 24 These tables show the length of the season in the different sections of the cotton belt. The winters in the cotton belt are seldom severe and the temperature rarely reaches zero except in the northern portion, where the climatic conditions are liable to extreme or sudden changes. The conditions are generally unfavorable to cotton production where the winter and spring temperatures are very low, as the growing season is liable to be too short for the maturity of the crop. Table 3 shows the minimum temperature of the three winter months in a number of localities in different sections of the cotton belt. Table 3. — Winter minimum temperature at stations of the cotton belt of the Southern States. Station. NOKTHEItN PORTION. Atlanta, Ga Charlotte, N. C Chattanooga, Tenn El Paso, Tex Fort Davis, Tex Fort Elliott, Tex Fort Smith, Ark Knox ville, Tenn Little Rock. Ark Memphis, Tenn Nashville, Tenn Length of record. Tears. 13 13 13 14 11 10 10 21 13 20 21 Mini- mum. °F. 2 5 7 5 3 14 7 16 5 8 10 Month and year. Jan., 1886 Dec, 1880 Jan., 1886 Dec, 1880 Jan., 1886 Jan., 1888 Jan., 1886 Jan., 1884 Jan., 1886 Jan., 1886 Jan., 1884 Mean minimum. Dec °F. ■67.6 35.5 35.6 32.9 33.2 25.6 33.8 32.3 38.5 38.1 33.9 Jan. °F. 35.4 33.8 34 30.7 30.1 18.7 26.8 30.6 33.7 32.8 30.5 Feb. Number of times minimum was zero and below. °F. 39.7 37.4 37.7 35.8 34.8 24.2 32.3 34.2 38 37.7 34.1 148 THE COTTON PLANT. Table 3. — Winter minimum temperature at stations of the cotton belt, etc. — Continued. Station. MIDDLE PORTION. Auburn, Ala Augusta, Ga Charleston, S. (J Green Springs, Ala Hatteras, N."C Kitty Hawk, N. C Montgomery, Ala Palestine, Tex Shreveport, La Union Springs, Ala Vicksburg, Miss Wilmington, N. C SOUTHERN PORTION. Brownsville, Tex Cedar Keys, Fla Galveston, Tex Indianola, Tex Jacksonville, Fla Mobile, Ala New Orleans, La Pensacola, Fla Kio Grande City, Tex San Antonio, Tex Savannah, Ga Length Mini- record. mum. Tears. o _p. 14 3 20 G 20 10 27 2 17 8 17 5 19 5 10 0 20 1 24 8 20 3 21 0 16 18 10 15.5 21 11 14 12 20 15 21 11 21 15 12 15 15 19 15 6 21 12 Mean minimum. Montli and year. Dec. «Tan., 1884 Jan., 1886 Jan., 1886 Jan., 1886 Dec., 1880 Feb., 1886 Jan., 1S86 Jan., 1886 Jan., 1886 Jan., 1886 Jan., 1886 Jan., 1884 )I)ec, 1880,- ^Jan., 18815 Jan., 1886 Jan., 1880 Jan., 1886 Jan., 1886 Jan., 1886 Jan., 1886 Jan., 1886 Jan., 1881 Jan., 1886 Jan., 1886 39. 7 30 44.8 42 40.1 40.0 42.6 41.6 42.7 39.9 51.4 51.5 50.1 Jan. Feb ° F. 38.2 38.8 44.5 39.3 36. 6 40.1 38.3 39. 9 38.9 50 51 47.4 43.8 47.5 43.6 47.3 46.3 47.7 41 43.7 o js\ 44.8 42 46.2 41. 9 39.4 44.4 43.8 43.1 44.2 41.5 55.2 54.9 52.9 49.9 50.8 47.6 51.2 51 54.2 46.8 46.9 Table 4 gives the mean monthly temperature of stations in the dif- ferent sections of the cotton belt during the growing season, Table 5 gives the mean minimum temperature, and Table C gives the mean maximum temperature of the same place. Table 4. — Mean monthly temperature of stations in the cotton belt. Station. NORTHERN SECTION. Atlanta, Ga Charlotte, N. C Chattanooga, Tenn El Paso, Tex Fort Davis, Tex Fort Elliott, Tex Fort Smith, Ark Knoxville, Tenn Little Pock, Ark Memphis, Tenn Nashville, Tenn Mean MIDDLE SECTION. Auburn, Ala Augusta, Ga Charleston, S. C Hatteras, N. C Kitty Hawk, N. C Montgomery, Ala Palestine, Tex Shreveport, La Vicksburg, Miss Wilmington, N. C Mean Apr. oF_ 61.6 59.9 62.7 64.3 60.7 57.1 62.8 58 63.5 62.5 59.6 61 63.4 64.6 64.4 56.9 56 65.8 66.7 67.3 65.9 60.5 63.2 68.9 68.8 68.9 73.1 68.4 69.7 68.9 66. 7 70.2 70.3 67.9 9.2 71.4 72.5 72.3 67.2 65.9 73.4 71.6 74 72.9 67.2 70.8 June. °F. 75. 7 75.9 75.7 81.5 78.3 73.7 77.2 74.2 77.8 78 75.9 76.7 76.7 79.2 79.4 74 74.2 80.4 79 81 79.5 76.2 July. °F. 78! 3 78.6 78.5 83.4 76.8 78.1 81.5 77.4 81.1 81.4 79.6 79.5 78 82.2 82 77.6 78.3 82,6 82.3 83.8 82.4 79.5 80.9 Aug. = F. 76.3 76.4 76.7 80.8 73.8 76.2 79 75.4 79.1 79 Sept. 78.1 80.2 80.8 77.8 76 80.2 81.3 82.4 80.9 78.4 79.6 °F. 7l!l 71.3 71.2 73.6 68.2 69.4 73.4 69.9 73.3 72.4 70.3 71.3 74 70.4 76.1 74.1 73.7 76 75.5 75.7 75.4 73.5 Oct. °F. 62' 64.6 61.6 63.8 61.9 58.1 62.8 58.8 63.7 62.7 60.3 61.8 64 64.8 67.2 65.1 64.1 65.5 66.8 66.4 65.9 62.1 Nov. 50.7 50.5 51.2 50.6 43.5 50.6 47.8 51.3 50.9 48.8 49.6 53.8 55 58.3 56.1 53.9 55.6 56.5 55.6 56 55.6 74.4 65.2 I 55.6 CLIMATOLOGY AND SOILS. 149 Table 4. — Mean monthly temperature of stations in the cotton belt — Continued. Station. SOUTHERN" SECTION". Brownsville, Tex Cedar Keys, Fla Galveston, Tex Jacksonville, Fla Mobile, Ala Xew Orleans, La Pensacola, Fla Rio Grande City, Tex. San Antonio, Tex. Mean . Apr. ' ° F. 74.4 69.8 70 68.9 07. 1 08.9 67.7 74.2 69.4 Savannah, Ga. 66.6 May ° F. 79.3 75.5 76.1 72 74.2 74.9 73.4 80.9 74.8 73.2 75.4 June. 80. 80. 79.8 85. 5 81.2 79.6 81.3 July. OF. 83.8 82 84.3 82.9 82.6 82.7 81.4 87.4 83.5 82.4 83.3 ° F. 83.1 81.8 83.5 81.7 80.7 81.8 80.9 86.3 82.6 79.1 Sept. ° F. 79.7 79.5 79.3 78.3 74.8 78.5 77.7 81.6 77.2 Oct. 82.1 0 F. 70.1 72.6 72.6 70.5 68.2 70.4 67.8 74.5 69.5 66.9 70.3 Nor. o jp_ 66.8 63.5 62.3 62.5 58.5 61 59.4 64.9 58.3 58.6 61.5 Table 5. — Mean maximum temperature of stations in the cotton belt. NORTHERN SECTION. Atlanta, Ga Charlotte, N. C Chattanooga, Tenn El Paso. Tex Fort Davis, Tex Fort Elliott, Tex Fort Smith, Ark Knoxville. Tenn Little Rock, Ark , Memphis, Tenn Nashville, Tenn Mean MIDDLE SECTION. Auburn, Ala Augusta, Ga Charleston, S. C Hatteras, N. C Kitty Hawk, N. C Montgomery, Ala Palestine, Tex Shre veport, La Vicksburg, Miss Wilmington, N. C Mean SOUTHERN SECTION. Brownsville, Tex Cedar Keys, Fla Galveston, Tex Jacksonville, Fla Mobile, Ala New Orleans, La Pensacola, Fla Rio Grande City, Tex San Antonio, Tex Savannah, Ga Mean Apr. °F. 71 70.4 74.2 80.1 75.1 70.9 74 68.3 73.3 71.5 69.5 72.6 72.7 75.3 71.9 63.2 62.7 76 76.3" 77.4 75.4 69.3 May. 72 82.4 76.1 75.1 78 75.5 76.1 74.6 87.8 79.7 74.7 78 °F. 78.3 78.9 79.6 88.9 82.9 77.3 80.3 77.6 79.5 79.4 78.2 80.1 80.9 83.4 79.5 72.5 72.9 83.8 81.2 84 82.9 76.6 June. July. °F. 84.' 3 85.4 85.3 97.6 90.4 85.6 87.6 84.1 86.7 °F. 86.' 7 87.9 87.9 90.6 92.6 87.1 90.1 90.1 89.1 87.2 84.8 89.1 86.4 79.2 80.8 89.8 88.4 91 84.8 91.8 88.9 82.6 84.5 91.9 92 93.7 91.7 86.7 79.8 86.2 81.9 81.5 83.6 83.2 82 80.1 91.5 84.7 81.4 90.2 86.4 87.4 89 88.8 87.2 86.1 96.7 91.2 87.5 83.6 91.3 87.9 89.6 91.1 90.8 89.1 87.9 99.2 94.1 90.1 91.1 Aug. Sept. OF. 84.3 85.2 85.6 94.4 83.3 87.8 89.9 85 88.1 °F. 78^8 79.9 80.3 86.7 79.6 81.9 84.5 79.9 82.2 81.2 79.7 81.3 89.3 87 81.9 80.1 88.6 91.3 92.4 90.2 85.9 87.3 91.3 87.9 88.7 89.7 88.2 88.2 87.6 98 93.2 87.8 82.8 87.6 86.1 84.1 85.6 85.7 84.8 84.6 92.1 86.7 82.9 Oct. OF. 70.5 71.3 71.1 78.2 74 70.3 74.5 69.5 73 71.3 70.3 Nov. OF. 60.4 60 59.6 64.9 62.8 56.3 61.4 57.4 60.1 59 57.8 72 00 82.7 74.7 63.8 84.6 75.2 65.4 82.2 74.1 65.6 78.8 70.2 61.7 79.1 69.6 59.9 85.3 75. 5 65.3 85.2 77.1 66.1 85.2 76.6 65.4 84.3 75.2 65.4 81.1 69.1 64.7 73.7 83.8 79.2 77.6 78.3 77 77.3 75.6 85.2 80.1 74.7 64.3 74.9 70.6 67.6 71.1 67.4 68.1 67.6 74.9 68.6 67.3 69. 150 THE COTTON PLANT. Table 6. — Mean minimum temperature of stations in the cotton belt. Station. NORTHERN SECTION. Atlanta, Ga Charlotte, H". C Chattanooga, Tenn El Paso, Tex Fort Davis, Tex Port Elliott, Tex Fort Sm ith, Ark Knoxville, Tenn Little Rock, Ark Memphis, Tenn Nashville, Tenn Mean MIDDLE SECTION. Auburn, Ala Augusta, Ga Charleston, S. C Hatteras, K. C Kitty Hawk, N. C Montgomery, Ala Palestine, Tex Shreveport, La Vicksburg, Miss Wilmington, N. C Mean SOUTHERN SECTION. Brownsville, Tex Cedar Keys, Fla Galveston, Tex Jacksonville, Fla Mobile, A la New Orleans, La Pensacola. Fla Eio Grande City, Tex San Antonio, Tex Savannah, Ga Mean Apr. °F. 52. 2 49.4 52.2 48.4 46.3 43.4 51.7 47.7 53.7 53.4 49.7 49.8 54 53.8 56. 9 50.6 49.3 55.6 57 57.3 56.7 51.7 54.8 66.5 63.4 64.9 59.7 58.7 61.8 60.8 60.5 59.2 57.4 61.3 May. °F. 59.6 58.7 58.2 57.3 54 52.1 57.6 55.8 60.9 61.1 57.5 57.5 61.9 61.6 65.1 61.8 58.9 63 62 64 63.4 57.7 61.9 72.3 69 70.7 60.4 65.1 67.7 66.7 70.3 64.8 65.1 67.2 Jinn °F. 67.2 66. 3 06.1 65.7 62.2 61.8 66.8 64.3 68.8 69 66. 5 65. 4 68.8 69.3 66.3 68.9 67.6 71.1 09.7 71 70.5 67.9 69.1 75.1 74.3 77.1 72.5 72.1 74.1 73.6 74.3 71.2 71.7 73.6 July, °F. 70 69.3 69.2 68.7 64. 3 65.7 70.3 67.4 72. 1 72.7 70. 1 69 71.2 72.6 69.3 72.6 72 73.3 72.6 73.9 73 72.2 Aug. °F. 68.4 67.6 67.8 67.2 62. 3 64.6 68.3 65.8 70 70.4 67.7 67. 3 69.4 71 74.5 72.7 72 71.7 71.3 72.3 71.6 70.9 72.2 76.3 77.5 79 74.6 74.4 76.2 74.8 75.6 72.9 74.6 71.1 74.9 75.6 78.2 73.7 73.2 75.4 74.2 74.6 72 70.4 75.0 Sept. °F. 03.4 62. 6 62 60.4 56.8 56.9 62.3 59.8 64.4 63.6 60.9 61.2 65.4 66.1 70 69. 4 68.3 66.7 65. 9 60.2 66.4. 65.8 Oct. OF. 53. 6 51.8 52.1 49.4 49.7 45.9 51.1 48.1 54. 3 54 50. 3 Nov. °F. 43. 4 41.4 41.4 37.6 38.3 30. 8 39. C 38.2 42.5 42.8 39.8 50. 9 67 71.9 72.9 74.4 71 63.9 72.1 70.9 71. 1 67.7 09 53. 4 54.4 00.2 60 58.6 55.6 56.4 56.2 56.5 55.1 56.6 39. 0 43.7 44.6 50.9 50.4 47.9 45.9 46.9 45.7 46.7 46.5 46.9 66.4 66 67.5 62.8 59.3 63.4 62.1 63.7 58.9 59.1 70.5 62.9 58.6 56.4 57 53.8 49.6 53.9 51.2 54.9 48.1 49.9 53. 3 These tables show a long season of uniformly warm conditions, especially in the middle and southern sections, which are very favorable to the production of cotton. Table 7 gives the average rainfall, with the number of rainy days and the number of clear days, for the month of May in the three sections of the cotton belt from 1871 to 1891, inclusive. The detailed informa- tion for the several stations is not given in the bulletin from ^vhich these tables are taken. CLIMATOLOGY AND SOILS. 151 Table 7. — Average rainfall, average number of rainy days, and average number of clear days for the month of May for the years 1871 to 1S91, inclusive. [These averages were obtained from data furnished by all regular stations throughout the cotton belt.] Northern cotton belt. Middle cotton belt. Southern cotton belt. Tear. Rainfall. Num- ber of rainy days. Num- ber of clear days. Rainfall. Num- ber of rainy days. Num- ber of clear days. Rainfall. Num- ber of rainy days. Num- ber of clear days. 1871 Inches. 4.66 3.62 5.56 1.06 2.97 5.94 1.53 3 3.52 4.22 3.34 5.89 3.34 5.20 4.35 2.83 3.68 3.92 2.51 4.05 2.67 12 10 12 6 12 10 6 11 8 8 11 11 11 12 7 12 12 7 13 6 8 11 4 13 10 9 14 11 13 13 10 10 15 11 11 13 11 10 15 14 13 Inches. 6.12 7.78 7.63 2.97 2.79 5.24 3 5.07 3.54 3.66 2.57 3.80 3.66 6.29 6.57 2.88 4.06 5.27 2.66 5.76 2.85 8 10 12 7 7 10 6 9 10 9 8 8 6 8 12 6 11 11 6 10 6 16 12 8 14 11 9 15 12 13 14 10 8 11 9 8 16 10 8 15 12 13 Inches. 4.30 2.78 8.86 2.94 3.20 3 1.67 4.37 3 4.86 2.40 4.40 4.32 5.24 5.40 2.08 3.90 4.54 1.08 4.24 1.92 9 6 13 6 7 6 6 9 6 10 7 8 7 10 11 4 9 10 4 10 4 9 1872 10 1873 7 1874 13 1875 12 1876 12 1877 11 1878 9 1879 14 1880 10 1881 11 1882 8 1883 11 1884 12 1885 7 1886 15 1887 10 1888 10 1889 15 1890 12 1891 13 Mean 3.71 10 11 4.48 9 12 3.74 8 11 Table 8 gives tlie normal precipitation, the average number of rainy days, of clear days, and of cloudy days at the several stations of the three different sections of the cotton belt for the months of June, July, August, and September, which, together with the month of May, are the most important months of the crop season. Table 8. — Precipitation for June, July, August, and September in the cotton belt. Normal precipitation. Average number of rainy days. June.! July. Aug. Sept. June. July. Aug. Sept. Average number of clear davs. June. July. Aug. Sept NORTHERN' SECTION. In. 4.41 4.67 4.37 53 Atlanta, Ga Charlotte, N. C Chattanooga, Tenn . . El Paso, Tex Tort Davis, Tex ' 2. 09 I Fort Elliott, Tex | 3.18 Fort Smith, Ark | 4.27 Knox ville, Tenn 4.29 Little Rock, Ark j 4. 39 Memphis, Tenn j 5 Nashville, Tenn 4. 34 Mean 3. 77 MIDDLE SECTION. Auburn. Ala Augusta, Ga Hatterae, N. C Kitty Hawk, N.C- Montgomery, Ala. Palestine, Tex Shreveport. La . . . Vicksburg, Miss . . Wilmington, N. C. 5.28 4.24 4.86 4.81 4.92 3.51 3.57 4.35 5.94 In. 4.63 5.86 3.72 2.17 3.31 2.32 4.09 4.37 3.88 3.29 4.54 3.83 4.37 5.17 6.33 5.81 4.24 2.82 3.68 4. 15 7.27 In. 4.39 5.46 4.16 1.87 4.17 3.27 3.65 4.28 3.92 3.82 3.62 In. 4.21 3.24 4.24 1.22 2.90 1.77 3.61 3.06 3.23 3.23 12 12 14.3 4 8.2 6.9 10 13.3 11.6 11.6 11.2 3.87 Mean 4.60 I 4.87 4.20 4.83 6.52 7.48 3.80 2.51 2. 05 3. 50 7.-80 3.29 3.74 6.61 5.18 3.11 3.21 4.25 3.85 6.70 10.2 11.3 10.6 10.8 12.3 7.9 8.4 10.9 12 4. 74 I 4. 43 10. 5 10 12.2 13.1 8.4 10 5.3 8.7 12.7 10 10 10.5 13.1 10.6 13.2 9.4 10.8 8.3 8.4 12.6 8.9 9.9 9.4 9.7 8.9 10.9 5.3 7.4 5.3 8.1 9.6 S. 4 8.6 9.3 8.7 8.8 19.5 15.7 13.2 11.6 8.4 10.1 9.2 6.4 8.9 13.5 15.4 14.5 13.9 10.2 11.3 11.6 9.1 8.6 15 14 16.1 15.3 10.3 14.5 13.7 11.2 12.6 10.5 11.3 16.7 13.7 17.1 14.9 13.5 13.4 13 11.4 10 10 11.2 11. 1 11.4 12 7. 1 8.7 11.3 15.1 10.4 I 8.3 -10.8 13.5 12.4 10.8 13.8 12.9 12 6.7 5.7 8.7 13.7 7.8 5.4 8.6 8.2 8.3 9.1 9.8 10.8 10.6 6.4 7.5 8.1 8.8 10.2 11.3 10.6 9.7 7.7 7.5 8.7 13.9 9.3 11.4 9.3 9.7 8.7 8.8 9.4 8.2 11.2 8.4 8.3 13 13.6 11.2 8.5 10.2 10.2 13.6 11.5 11.7 12.9 13.9 12 10.4 11.6 152 THE COTTON PLANT. Table 8. — Precipitation for June, July, August, and September, etc. — Continued. Station. SOUTHERN SECTION. Brownsville, Tex Cedar Keys, Pla Galveston, Tex Jacksonville, Fla Mobile, Ala New Orleans, La Pensaeola, Fla Eio Grande City, Tex. San Antonio, Tex Savannah, Ga Mean Normal precipitation. June. July. Aug. Sept. In. 3.25 G. 83 5.76 0.03 6.06 6. 66 5.85 2.27 2.46 6.75 5.18 In. 2.22 8.68 3. 04 6.36 6.60 6. 38 6.55 1.29 2.68 5.34 la. 3. 00 7.72 2. 04 6.80 6.41 6.02 8.13 2.94 3.45 7.65 5.59 In. 7.73 5.37 7.07 8.06 5.06 4.82 5.25 3. 78 4.16 5.89 5.71 Average number of rainy days. June. July. Aug. Sept, 6.9 11.3 7.2 14.8 12.9 13.8 12.4 4.8 6.3 13.1 10.3 4.5 14.5 8.6 16.7 15.1 15.8 14.7 3.6 6 12!8 11.2 7.3 13.1 10.3 15.5 12.8 14.5 13.6 4.9 0.7 13.5 11.2 11.5 10.1 11 14.1 9.5 10.6 10 9.4 9.7 11.5 10.7 Average number of clear days. June. July. Aug. Sept 13. 5 6.2 11.7 7.7 8.1 8.1 10 14.1 8.4 7.3 9.5 15. 1 7.7 13.2 8.8 7 8 9.1 19 10.8 7.9 10.6 10 12.4 8.7 13. 3 8.7 8.1 8.1 10.5 14.6 8.7 7.7 11.1 11.9 12.6 9.1 11.2 10.9 J 2. 4 11.5 10.2 8.7 10.9 Station. NORTHERN SECTION. Atlanta, Ga Charlotte, N. C Chattanooga, Tenn El Paso, Tex Fort Davis, Tex Fort Elliott, Tex Fort Smith. Ark Knoxville, Tenn Little Rock, Ark Memphis, Tenn Nashville, Tenn Mean . . MIDDLE SECTION. Auburn, Ala Augusta, Ga Hatteras, N. C Kitty Hawk, N. C Montgomery, Ala Palestine, Tex Shreveport, La Vicksburg, Miss Wilmington, N. C Mean SOUTHERN SECTION. Brownsville, Tex Cedar Keys, Fla Galveston, Tex Jacksonville, Fla Mobile, Ala New Orleans, La Pensaeola. Fla Eio Grande City, Tex San Antonio, Tex Savannah, Ga Mean Average number of partly cloudy days. June. July. Aug. Sept 14.6 13.3 14.4 9.5 12.5 12.8 13 15 14.6 14.7 17.1 14.6 14 14.7 6 13.5 13.5 16.1 15.7 15 14.4 13.6 12.5 16.6 13.8 15 15 15.5 14. 6 12.1 16.1 14.9 14.9 14.7 14.6 14.8 11.6 13.1 12.2 14.5 14.1 13.8 16.1 14 16.6 14.6 13.4 15.3 15.8 13.6 15.1 15.1 14.4 14.8 12.8 14 14.1 16.1 16.3 17.4 15.8 8.7 16.2 16.2 14. 15 12.3 15.4 11.7 12.6 9.9 9.6 14 12.4 11.7 14.2 12.6 15 15 11.8 14.9 16.3 14.8 13.9 15. I 13.5 14.4 15.1 13.9 12.9 16.4 15.9 17.9 14 10.7 18.3 15.3 15 11.7 10.8 12.1 8.9 13.3 12.1 11.2 11.3 10.7 10.5 9.7 10 10.9 11 12.8 12.9 10.7 12.2 12 13.1 12.3 12.1 11.8 11.7 12.1 Average number of cloudy days. 7.4 8 6.1 1 1.8 3.7 5.4 6.6 5.3 6.1 6.5 9.6 7.2 13.8 5.9 8.8 5 5 5.7 7.9 7.6 July. 7.3 8.5 6.7 2.7 4 3.4 4.9 6.3 5.6 5.6 6 6.9 7.6 6.3 6 7.7 3.5 4.5 6.2 7.8 4 7.2 4.5 7.3 6.9 6.4 5.4 2.5 9.3 3.7 6.1 7.7 5.6 6.1 3.3 4 6.9 Aug. 9.6 7 4.3 4.4 5 6.1 '6.7 4.1 5.6 5.6 6.6 7.8 8 7.7 6.4 3.2 3.5 4.7 9 3.5 8.4 4.8 5.9 7 5 6.5 5.7 4 8 Sept. 5.7 8.7 6.6 4.4 5.8 3.7 6.5 7 5 7 7 7.8 7.7 6.2 7.2 7.6 6.6 6.4 6.1 5.2 6.7 8.7. 6.8 6 5.3 6.4 8 9.6 6.8 Little comment is here made upon the data contained in these tables. They are but broad, general statements of the climatic conditions, which can be used for general comparisons and as a basis for other more detailed work. CLIMATOLOGY AND SOILS. 153 SOIL. Cotton is at present cultivated with more or less success on nearly all kinds of soils within the region in which the climatic conditions are favorable to its growth and development. It is grown alike on light sandy soils, on loams, on heavy clay soils, and on bottom lands, but not with equal success on all of these different types of soil. On the sandy uplands the yield of cotton is usually very small; on clay lands, especially in wet seasons, the plants attain large size, but yield a small amount of lint in proportion to the size of the plants. This is also likely to be the case on bottom lands. The safest soils for the crop are medium grades of loam. On the bottom lands in very favorable sea- sons the crop often produces a very large yield, but it is not so cer- tain, and in unfavorable seasons the plants are liable to disease and to insect ravages. It becomes important for the cotton planter to understand the soils of his farm in order to get the greatest possible advantage from the soils adapted to cotton and to appreciate the fact that he can not hope to compete successfully in the production of cotton on other kinds of land. It is the purpose of this chapter to summarize our present knowledge of the relations of soils to crops from the investigations which have been carried on in this country and , abroad, not only to concentrate the light upon this subject for immediate benefit in its application to the problems of present interest to the cotton planter, but to stimulate interest in this subject, which presents a very promis- ing field for investigation. The study of soils is naturally divided into two parts — (1) the soil considered merely as a source of food supply for plants and (2) the study of the physical conditions in the soils, especially of moisture and heat, which are equally essential for the growth and development of plants. CHEMICAL PROPERTIES OF SOILS. The chemistry of soils has been very earnestly studied in the past fifty years. It was once very generally believed by agricultural chemists that the chemical analyses of any particular soil and plant would show •what might be lacking in the soil for the production of a normal crop of the plant. Later investigations have dispelled this idea, and have shown that there is no simple relation between the chemical analysis of a soil and the crop-producing power. This is especially marked in the case of nitrogen, phosphoric acid, potash, and lime, which experi- ence has taught us to apply to the soils in the form of fertilizers. That these fertilizers have been of great and undoubted value in the increased production of crops no one can for a moment doubt, but the effect of these fertilizers on the soil in increasing the yield of crops, and the real principle upon which we should apply them to the land is a subject of 154 THE COTTON PLANT. considerable doubt and a matter of very diverse opinions among scien- tific men. Soils, to a depth of 1 foot, rarely contain less than 0.05 per cent each, of phosphoric acid, potash, and lime, which corresponds to about 1 ton of these substances per acre, and they usually contain from two to twenty times this amount. A soil, however, containing as much as 10 tons of phosphoric acid or of potash or of lime per acre may be natu- rally unproductive or may be readily exhausted by injudicious methods of cropping and cultivation. Crops remove but a very small amount of mineral food — so little, indeed, that the loss can not be detected by the most careful chemical analysis of the soil after a large crop has been removed. It has been repeatedly shown that very barren soils often contain as much plant food as others which are considered fertile. It is strange indeed that the application of a fertilizer containing no more than 20 pounds per acre of phosphoric acid or of potash to a soil which, to a depth of 1 foot, already contains from 2,000 to 40,000 pounds, should make the difference between a poor and a large crop. It is no unusual thing to find in the experience of farmers and in the careful experiments conducted at the experiment stations that the increased crop due to this addition of fertilizing material contains more plant food than is contained in the fertilizer which has caused the increase. It is commonly believed now by agricultural chemists that the reason small quantities of fertilizers have such an effect upon soils containing such enormous quantities of these same ingredients is that only a very small portion of the plant food in the soil is in a condition to be readily available to plants, as by far the larger proportion of the plant food is in the form of minerals which the plants can not readily assimilate. Eecent investigations have therefore been turned to the considera- tion of the amount of food material in the soil which is readily avail- able to plants. Two methods have given considerable promise of valuable suggestion as to the amount of available plant food in soils. One is based upon the assumption that the available plant food is com- bined with the partially decomposed organic matter or humus in the soil, and that if this humus is extracted and an analysis made of it the results will show tfhe amount of the food material in the soil readily' available to plants. This idea has been strongly advocated by Grau- deau and has been developed in this country with valuable results by Hilgard1 and Snyder.2 For years an effort has been made to find a solvent of such a char- acter and concentration as shall dissolve out of the soil itself about the same amount of plant food that a plant would extract through its roots during its season of growth. Dyer3 has made a very thorough and 1 California Sta. Ept. 1891-92, p. 241. 2 Minnesota Sta. Bui. 41. 3 Jour. Chem. Soc, 1894, p. 115 (E. S. E., 5, p. 1013). CLIMATOLOGY AND SOILS. 155 systematic study of the soils of some of the experimental plats at Rothamsted, and has obtained some exceedingly interesting and prom- ising results. He used as a solvent a 1 per cent solution of citric acid, which lias about the same acidity as the juices expressed from the roots of a large number of plants which he examined. Citric acid was selected because it is an organic acid very commonly xuesent in plants, and is easjT to obtain in a comparatively pure state. This dilute acid was allowed to act on the soil for a considerable time at a certain temperature, the soil and solvent being occasionally shaken. The amount of plant food extracted in this way from the soil was estimated, and was assumed to be the amount of food material in the soils available to plants. Very striking relations were shown to exist between the matters extracted by this process and the crop yields from the different plats. A great deal of light has been thrown upon the chemical constitution of soils by these investigations, but the point is not reached at which the kind of crop best adapted to a given soil or the additions of plant food necessary to make the soil suitable for crops to which it is not nat- urally adapted can be predicted with certainty from a chemical analysis of soil and crop. This is undoubtedly due in large part to the fact that the agricultural chemist has paid too little attention to the other con- ditions of growth, especially to the physical conditions of the soil, which are of equal importance with the chemical composition of the soil in determining crop production. A very large amount of work has been done in this country in the investigation of the chemical composition of soils of the cotton region. Very valuable and suggestive work has been published in the reports of the Kentuckj^ Geological Survey, in the reports of Professor Hil- gard on the agriculture of Mississippi and California, and especially in the elaborate summary of the work of Hilgard and others in the Tenth Census.1 This latter report describes in great detail the soils of all of the principal agricultural regions of the cotton belt, and gives a great many analyses showing their chemical composition as determined by digestion with strong hydrochloric acid. (See p. 164.) The limits of this chapter permit of only the most general statements as to the results of the vast amount of work done prior to the publica- tion of the Tenth Census and of the special investigations undertaken for the Tenth Census itself on the chemical composition of soils adapted to cotton. Hilgard states,2 as a result of his investigations and that of his colaborers in this exhaustive research, that as a rule the relative proportions of phosphoric acid and of lime seems to govern the produc- tiveness of our virgin soils. A soil containing only 0.05 per cent of phosphoric acid must be regarded as seriously deficient in this element unless it contains a large amount of lime. In sandy loam soils 0.1 'Tenth Census of the U. S., 1880, Vols. V and VI, Cotton Production, parts 1 and 2. 2 Tenth Census of the U. S., 1880, Vol. V, Cotton Production, part 1, p. 78. 156 THE COTTON PLANT. per cent, when accompanied by a fair supply of lime, gives a fair pro- ductiveness for from eight to fifteen years. With a deficiency of lime, however, twice that amount will only serve for a similar time. The maximum amount of phosphoric acid found in a pine upland soil by his method of analysis is about 0.25 per cent in the splendid table-land soils of west Tennessee and Mississippi. In the best bottom soils, or buckshot soils, of the Mississippi Valley, 0.3; in that of the black prairie of Texas, 0.46, and in a red-clay soil from Tennessee, 0.563 per cent. This latter figure would imply the presence of 22,000 pounds of phosphoric acid per acre to a depth of 1 foot, soluble in the acid of the strength used in the chemical analyses. Virgin soils containing less than 0.06 per cent of potash may be usually assumed to be deficient in available potash, and an application of potash to such soils may be expected to give beneficial results. Sometimes, however, a soil very rich in lime and phosphoric acid shows good productiveness despite a very low potash percentage. The amount of potash in heavy clay upland soil and clay loams ranges from about 0.8 to 0.5 per cent, in lighter loams from 0.45 to 0.30, in sandy loams it falls below 0.3, and in sandy soils of great depth it may fall below 0.1 per cent and still give good productiveness and durability, depending somewhat upon the amount of lime and phosphoric acid within the soil. It will thus be seen that the percentage of potash varies somewhat with the amount of clay. The buckshot soil of the Mississippi bottom contains 1.3 per cent of potash and 1.4 per cent of lime and is jet black with humus and may well serve as the type of a fertile soil. Hilgard states that in his experience few of the soils of the South contain less potash than the quantities above reported and that potash manures are not usually necessary for the exhausted soils of the South „ He also states that the universal preference given to phosphoric and nitrogenous fertilizers in the West and South is in accord with this inference. Hilgard states as a result of his experience that 0.1 per cent of lime in the lightest sandy soils gives character to the native timber growth. In clay loams there should be not less than 0.5 per cent, and the amount may rise advantageously to 1 or even 2 per cent. These conclusions of Hilgard as the result of the extensive inves- tigations for the Tenth Census still stand, and are likely to stand for a long time to come, as the standards by which we will compare the chemical composition of the soils in this country. They are very gen- eral, but it is not possible at present to give more specific statements. But little work has been done upon the chemical composition of the soils of the cotton belt since the completion of the investigations made for the Tenth Census. More recently Pagnoul1 states as a result of his own experiments, and from the experiments of Heherain, Joulie, and Dellisse, that a soil Carres Arables du Pas-de-Calais, Arras, 1894 (E. S. R., 6, p. 118). CLIMATOLOGY AND SOILS. 157 containing less than 0.1 per cent of phosphoric acid will probably be deficient in that substance and will be benefited by the application of phosphates. Miintz points out that this limit should not be arbitrarily fixed, as it would depend largely upon the form of combination of the phosphoric acid in the soil. In regard to the form in which phosphoric acid should be applied, it is stated that if the soil is calcareous super- phosphates can be used with advantage; if the soil contains less than 0.1 per cent of lime, superphosphates are apt to be injurious and a phos- phate containing free lime, such as Thomas slag, should be used. Precipitated phosphates are well adapted to the soils which need phos- phoric acid, but produce their best results on soils which contain smaller amounts of lime. Natural phosphates are used to the best advantage on humus soils when applied in a very finely ground condition. The author states as a result of his investigations that a soil containing 0.25 per cent of potash is not likely to respond profit- ably to an application of potash salts. . The normal proportion of nitrogen in good soils is stated to be about 0.1 per cent. Whether soils containing as much nitrogen as this will be benefited by an application of nitrogenous fertilizers will depend largely upon the condition of the nitrogen in the soil and upon the demands of the plant. When the amount of nitrogen is below this limit nitrogenous fertilizers are considered almost always indispensable. These views as to the chemical composition of soils and their relation to plant growth represent the views held at present both in this conn- try and abroad. They are the results of a vast amount of chemical investigation. It will be seen that they are very general and that no specific deductions can be drawn from the chemical analysis of a soil alone. PHYSICAL STRUCTURE OF SOILS. In the preceding section the results of the chemical investigations of sods have been given as these bear upon crop production. It has been shown in the first place that the total amount of plant foods in the soil has no simple or direct bearing upon the relation of soils to crop pro- duction. It is very generally believed, however, as a result of the investigations in this country and abroad, that fertile soils should have a certain quantity of each of these plant foods as determined by the extraction of the soil by strong acids. If the amount falls below the limits which have been assigned, experience has shown that in most cases the soils will respond to fertilizers or manures containing the elements which are thus shown to be deficient in the soil. But occasion- ally very poor, worn out, and exhausted soils contain as much of these plant foods as would correspond with the arbitrary limits which have been set for soils of average fertility. It has been stated likewise that promising investigations are being carried on with different solvents, and that it is believed by many that it will be possible to find a solvent 158 THE COTTON PLANT. of a nature and strength which will indicate very clearly the amount of plant food in the soil which is available to plants. We come now to a consideration of the physical texture and struc- ture of soils and especially to the conditions of moisture they maintain, which have a most important bearing- upon crop production, and which undoubtedly are often of such influence that they completely over- shadow in importance the chemical composition of the soil. The classification of soils into sandy, sandy loam, loam, clay loam, clay, and bottom lauds has a very distinctive and important meaning to the farmer, for he recognizes certain properties characteristic of each of these classes of soils and has learned from experience to expect certain results. On the light, sandy uplands the yield per acre is almost invariably small; on stiff clay lauds and on bottom lands, especially in wet sea- sons, the plants are inclined to make an excessive growth of the leafy parts and to put on little fruit in proportion to the size of the plants and the crops to be late in maturing. It is also a recognized fact and a matter. of wide experience that plants growing on some of these typical soils are much more subject to diseases and to insect ravages under unfavorable climatic conditions than on other soils. The safest soil for cotton is found to be a medium loam, although, as stated elsewhere, in favorable seasons and under favorable conditions very large yields are obtained from the heavier clay soils and on bottom lands. The texture of the soils, therefore, or the relative amount of sand and clay, has an important bearing on crop production, as has been shown in the experience of farmers throughout the cotton belt. It remains to be seen what these differences are which are dependent upon the relative amount of sand and clay, which the farmer can recognize at a glance, as he can not recognize differences in the chemical composition of the soil. The first thing which is very apparent in the consideration of this subject is that the character of the season, and especially of the amount and distribution of the rainfall, has a very marked and important influ- ence upon the yield of cotton on all of these different types of soil. It is no unusual thing for a cotton crop on a heavy clay or rich bottom land to be so badly diseased or to be so injured by insect ravages as to cause an entire failure of the crop, or to be so delayed in maturing, by reason of unfavorable climatic conditions and rank vegetative growth, that the crop does not mature before frost. In more favorable seasons the yield from these same lands may be enormous. Every farmer also recognizes when he puts in his crop the great uncertainty as to the yield, and it is no uncommon thing for the yield of cotton for an entire State to be double one year what it was the previous year. It is likewise a matter of very common experience that the cotton in one field or tract of land will be much more affected by unfavorable CLIMATOLOGY AND SOILS. 159 climatic conditions than the crop in an adjacent field of perhaps a different character of soil. We see here an indication of the great influence the conditions of rainfall and moisture have upon the cotton crop, and we have indica- tions also that these different types of soil maintain very different con- ditions of moisture for the plants. It may be said, in fact, that while climatic conditions determine the general distribution of plants — make it impracticable, for example, to grow cotton in a northern latitude — the texture of the soils, or the relative amount of sand and clay they contain, and the relation of these differently textured soils to moisture largely determine the local distribution of plants and explain why some soils are adapted to cotton, wheat, tobacco, or to truck farming, and why other soils are not so well adapted to these crops. As a rule, the relative amount of moisture maintained by different soils for crops under normal conditions depends upon the resistance the soil offers to the descent of the rainfall. The actual resistance, and therefore the relative amount of moisture, maintained by different soils depends upon the amount of space in the soils for the water to enter; upon the number of grains of sand, silt, and clay, for this will determine how much the space is divided up; upon how these grains are arranged, for this will have an influence upon the resistance or the friction the soil offers to the descent of the rain- fall; upon the amount of organic matter in the soil, and upon the depth of the soil. Soils contain, as a rule, about 50 per cent by volume of empty space; that is, in 1 cubic foot of soil there will be one-half cubic foot of space into which water or air can enter. In a sandy soil this space will not be divided up so much as in a clay soil; the sand having fewer grains the spaces between the grains are larger, so there is less friction and the water moves downward more quickly. These sandy soils will not, therefore, maintain so much moisture for the plants. The particles of clay soils, on the other hand, are so exceedingly minute, and there are such a vast number of them in the soil, that the spaces between them are exceedingly small and offer a great resistance to the descent of rain, so that the water moves very slowly and a large amount is maintained for the plants. A strong clay soil will usually contain three or four times as much water as a sandy soil, and this has a very important effect upon the growth of cotton. Advantage is taken of this fact in greenhouses, but in nature, of course, we can not control the conditions. In the latter case we must take the amount of rain which falls, but our different soils being so different in texture and in the resistance which they offer to the descent of the rainfall, maintain quite as different conditions of moisture as are supplied in practice under artificial conditions of green- house culture. Light sandy soils, being moderately dry, force plants to an early maturity, and these soils are used at present for the very 160 THE COTTON PLANT. profitable truck farming which has developed into such an important industry in recent years. Such soils are not adapted to wheat or grass, not necessarily because they are deficient in any particular food required for these crops, but because the dry conditions force the plants to an early maturity and the yield per acre is small. Clay soils, on the other hand, are adapted to wheat and grass, because they maintain uniformly moist conditions and the plants have a slow and prolonged growing period before it is time to mature seed. Clay soils are not adapted to early truck, because the crops are so late in maturing that they lose the advantage of the high market prices. The resistance or the friction in the soil determines the proportion of the rainfall which will be held back for the use of plants, and deter- mines largely the amount of moisture which different soils maintain. There must, however, be some automatic power to move this water from place to place in the soil and deliver it to the roots of plants as it is needed. This power exists in the soil in the force of surface tension, or, as it is more commonly called, capillary force, which may move water m any direction in the soil, either up or down or laterally. Fertilizers have an important effect upon this force of surface ten- sion. Lime, kainit, salt, plaster, and acid phosphate increase the sur- face tension and therefore increase the force which moves water to the plant. This probably explains many facts commonly met with in practice. A cotton soil should maintain very uniform conditions of moisture, for any marked or sudden variation, especially during the growing period, is apt to affect the vitality of the plant and have a marked effect upon the development of the crop. During the early growing season of the plant, up to the first of August, the soil should be con- tinuously moist, but not wet. A sandy soil, as a rule, is not sufficiently retentive of moisture, and the supply of moisture is so inadequate that the plants are small and are forced to an early maturity before they have gathered sufficient food material for a normal crop. On the other hand, a clay soil or a bottom land is liable to maintain too much mois- ture, and the plant takes on an excessive growth. If this condition is checked at the proper time and the plant is induced to mature fruit, the yield may be very large, but if this condition continues and the soil remains continuously moist after the first of August, the plants develop in a very luxuriant way, but with little tendency to put on fruit. Such soils may be greatly benefited by underdrainage, whereby the excess of water is artificially removed from the soil. This exces- sive growth may be checked also by fertilizers, especially by heavy application of phosphoric acid, which has a tendency to check the vegetative growth and hasten the maturity of the plant. The safest soil for the cotton crop is a deep loam, naturally well drained, but sufficiently retentive of moisture to maintaiu a uniform supply throughout the entire growing season. CLIMATOLOGY AND SOILS. 161 The following* is Hilgard's description of the famous buckshot soil of Louisiana, which is justly claimed to he the finest type of cotton soil in this country : The buckshot soil, in its store of plant food of all kinds, stands preeminent above all of the rest of the soils and well deserves its reputation of being the most pro- ductive and tillable soil iu the great bottom. Unlike many other clay soils, it may be tilled at any time when the plow can be propelled through it, because on drying it crumbles into a loose mass of better tilth than many elaborately tilled upland soils. It is so deep that the deepest tillage, even by a steam plow, would not reach beyond the true soil material. Its high absorptive power secures crops against injury from drought. Two bales of lint per acre can be produced on these soils with fair cultivation and good seasons. In South Carolina the ridge lands and the soils of the lied Hill forma- tion, covering an extensive area in the central and eastern portions of the State, are types of the most productive and most certain cotton soils, under good treatment, of any in the State. These soils, which may be considered a type of the finest cotton soils of that locality, contain from 25 to 30 per cent of clay and 40 per cent of silt. In the North this would make a very fair quality of wheat land. These soils maintain on an average during the growing season about 10 or 12 per cent of moisture for the cotton crop. The Sea Island cotton is best adapted to a very different kind of soil. The best soils for this variety are light, fine-grained, sandy soils, con- taining from 1 to 8 per cent of clay, from 1 to 0 per cent of silt, and from 75 to 90 per cent of fine sand. Soils of this character from James Island maintained during two growing seasons about 5 per cent of moisture and are very different from the best type of soils adapted to the upland cotton. The foregoing is a concise statement of the present views of the relation of climatic conditions and of soils to cotton production rather than an attempt to describe the individual soil formations. In the cot- ton report of the Tenth Census the principal soil formations of the cotton-producing States are very elaborately described, and form the basis for very valuable investigations.1 It is believed that this discussion of the principles will be of more value than a description of Jlie agricultural features of the different soils of the cotton belt, as it is more suggestive both to the practical farmer in his study of the best methods of dealing with his land and to the investigator who wishes to continue the study of the important problem presented in the relation of climate and soils to cotton production. Comparatively little work has been done in this country upon the physical properties of the soils of the cotton belt. Quite a number of mechanical analyses of soils were made by Hilgard's method for the report on cotton for the Tenth Census, but there is little attempt at an interpretation of these results, and in all that mass of literature 1 See also chapter on culture, p. 225. 1993— No. 33 11 162 THE COTTON PLANT. very little attention lias been given to the physical properties of the soils. Very little work has been done since then upon the study of the structure or physical properties of the soils of the cotton belt. TYPICAL SOILS OP THE COTTON BELT. As already noted, chemical examination of the soils of the cotton region has been more complete than physical studies. The following tables, compiled from Ililgard's report in the Tenth Census, referred to above, give the chemical analyses of representative samples of subsoil from the typical soil areas in all the States in which cotton is exten- sively grown. For a complete compilation of analyses the reader is referred to the original report. Classification of the typical soil areas of the cotton States. [The numbers refer to samples in the accompanying tables of analyses.] Metamorphic region : Red lands. Gray lands. Coal Measures region. Tennessee Valley region : The barrens. The red valley lands. ALABAMA. Middle division. Coosa Valley region : Flatwoods. Brown-loam and red-clay lands. Gray cherty lands. NortJi ern divi sio n . Tennessee Valley region — Continued. Sandy land of the Little Mountain range. Oak and pine uplands region: Oak and hickory uplands, with short- leaf pine. Gravelly hills, with long-leaf pine (3, 9, 20). Oak and hickory uplands, with long- leaf pine (140, 94, 92). Central or upper prairie region : Black prairies or canebrake (30, 16). So ii them division. Central or upper prairie region — Cont'd. Hill prairies (13). Blue-marl lands. Post- oak flatwoods region (25, 17). Lime hills or lower prairie region (139, 90). Long-leaf pine region (88). Alluvial region. 284, 441, Alluvial lands: Mississippi and St. Francis alluvial region (219). Arkansas River bottom (275, 390, 427, 416). White River lands (246, 384, 233). ' Saline River lands (337). Red River bottom lands (359). Poplar ridge lands east of White River (222, 447, 434). Gray-silt prairies of eastern Arkansas (323, 434, 468). Yellow loam and sandy pine-hills region : Gray sandy (350). ARKANSAS. Black prairie of the Southwest (328). Red-loam region, rocky, and hill lands : Gray and red loam timbered lands (314, 393, 317, 302). Western and central red-loam prairie region (353). Northwestern red prairie (254). Metamorphic soils (402). Northern barrens and hills region : Siliceous lands of chert, sandstone, and limestone (287, 308, 293, 242). Barrens and cherty magnesian lime- stone lands (266, 257,250). CLIMATOLOGY AND SOILS. 163 GEORGIA. Northwest Georgia : Gray sandy lauds of the metamorphic border. Flatwoods. Red-clay lands. Gray siliceous soils of the ridges. Brown and red loams (517). Yellow-clay lands. Sandy table or mountain lands. Alluvial lands. Metamorphic or mineral region (Blue Ridge) : Gray Bandy lands (212, 82). Red lands. Gray granitic lands (288). Metamorphic or mineral region (Blue Ridge) — Continued. Flatwoods. Central cotton belt : Sand and pine hills or border region. Oak, hickory, and pine uplands (165, 182). Red-clay hills. Long-leaf pine and wire-grass region: Lime-siuk or clay lands (500). Sandy pine barrens (509). Pine and palmetto flats. Coast region : Savanna lands. Live-oak and coast lands (511). LOUISIANA. Alluvial region : Alluvial region north of Red River (238). Red River bottom region (39). Alluvial region south of Red River. Tidewater region. Marsh lands. Blutf region. Attakapas prairie region : Black calcareous prairie (230). Brown-loam prairie. Gray silt, or pine prairies (195). Long-leaf pine region : Pine llats. Pine hills (134). Anacoco prairie. Central prairie region. Oak-upland region : Red lands (231, 184). Brown loam. Pale loam (232). Pine flats. • The northeastern prairie region: Rotten limestoue prairie region — Black prairie (172, 125). Black-jack prairie. Bald prairie (175). Ridge soils (141). Hickory hummocks (273). Sandy upland ridges. Bottom soils (177). The Pontotoc ridge. The flatwoods region: Post-oak flatwoods (230). White-oak flatwoods (147). Yellow-loam or oak upland region: Flatwood hills (119). Short-leaf pine and oak uplands ( 142 ) . Red lands (246). Sandy oak uplands (228), Brown-loam table lands (216). Bottom soils of the yellow-loam re- gion (156). MISSISSIPPI. The alluvial region of the Mississippi : Yazoo basin. Dogwood ridge (396). Sunflower basin (376). Deer Creek region (390). The cane-hills region : Bottom or valley soils of the region (117). Oak uplands belt (237, 116). Central prairie region : Soils of the central prairie region — Black prairie soils (188,210,44,40). Gypseous and hog- wallow prairie soils (242, 38). Sandy ridge lands. Marls of the central prairie region. Long-leaf pine region : Long-leaf pine hills (205, 292, 361, 181). Pine-flats region. Coast marshes (214, 88, 90). NORTH CAROLINA. Seaboard region (21, 25). I Long-leaf pine region — Continued. Long-leaf pine region: Pine flats (11). Sandy pine barrens. i Oak upland region (2). Level and rolling upland pine woods I Transmontane region. (13, 17, 37). 164 THE COTTON PLANT. Coast region. Lower pi7io belt. Upper pine belt. Red bills. Mississippi bottom. Upland of west Tennessee: Bluff region (16,17). Brown, loam table-lands (235). Western valley of Tennessee River. SOUTH CAROLINA. Sand bills Metamorphic regoin. Piedmont region. TENNESSEE. Highland rim. Central basin: Central limestone soils (2, 8). Orthis limestone binds (12, 14). Mulatto lands (10). TEXAS. The timbered upland: Short-leaf pine (1). The red hills (4, 6). Oak and hickory uplands (3). Prairies of the timbered region — Sandy prairies (7). Brown-loam prairies (34). Long-leaf pine region. The cross timbers (upper and lower). Southern coast prairies. East of the Brazos (9, 10). West of the Brazos (11). Southwest prairie and sandy deserts. Central black prairie region : Hog-wallow. Black sandy prairies (12). Central black prairie region — Cont'd. Black waxy prairies (15, 14). Northwestern red-loam region (16). Western and northwestern region : Gypsum region. Llano estacado or staked plain. Mountainous region of west Texas. Alluvial or river lands : Red River lands (40). Sabine and Trinity river lands. Brazos River lands ("sugar bowl") (19). Colorado, San Saba, Guadalupe, San Antonio, Neuces, and Rio Grande river lands. Analyses of typical soils of the cotton States. No. 140 94 92 139 90 Kind of soil. ALABAMA. Oak and pine uplands. Upland pine-woods soil . . . Alabama River hummock soil. Warrior River hummock soil. Lime-bills soil Sandy upland soil (cult.).. Alabama River second- bottom soil. Central prairie region. Black prairie soil do Upland sandy loam soil . Locality. Prattville Montgomery . Tuscaloosa . . ~ 0Q ■■> cS A 1* s> P.0 A 3 e 2-m <* c8 :_■ R In. 8 8| 1 Wilcox County .. Barbour County . Wilcox County . . Livingston Uniou Springs. do Post-oak flatwoods region. Post-oak flatwood soil. Post-oak prairie soil. . . Lime-hill region. Black shell prairie soil Upland brown-loam soil. . Upland pine- woods soil Livingston 10-48 Union Springs 12 Washington County . GosDort Andalusia Per cent. 1.39 6.04 1.13 1.91 3.27 17. 92 10.43 1.88 10.05 11.98 1.75 3.09 1.58 Per cent. 0.00 4.84 Per cent. 0.74 5.36 5. 18 4. 40 Per cent. 0.04 .34 .25 7.77 1.39 2.29 12.42 11.49 1.26 10.20 6.02 5.16 3.40 3.02 5.55 I .60 I. 1.78 6.94 7.86 6.54 6.98 5.42 2 1.14 .44 .29 .21 .55 .14 .17 Per cent. 0.06 .14 .27 .23 .03 .20 Per cent. 0.07 .14 .47 .18 .01 .22 1.96 .98 .08 .20 .37 . 37 29. 20 .12 | .10 .11 I .09 CLIMATOLOGY AND SOILS. 165 Analyses of typical soils of the cotton States — Continued. Kind of soil. ARKANSAS. Alluvial lands. Black bottom Arkansas bottom waste.. . Sandy loam Arkansas bottom Stiff chocolate-colored soil Arkansas River bottom cotton soil. Oil trough bottom soil Bottom soil Sandy loam Cache River bott om Second bottom soil Stiff Red River bottom. .. Sandy soil ■ Hill iand Little Prairie soil Locality. ££ 5;i — - = P Gray silt prairies of eastern Arkansas. 323 Soil of Grand Prairie. 434 i Soil of Little Prairie . 408 | Gray silt prairie Yelloiv loam and sandy pine hills region. 350 Pine hills soil. 328 j Black calcareous prairies . Red-loam region, rocky and hill lands. Sandy soil Brownish-gray soil Gray sandy soil Red ferruginous soil Red sumac prairie Prairie soil Granite soil Northern barrens and hills region. Sandy soil Brush Creek barrens Brownish-colored soils do Barrens soil Upland siliceous soil GEORGIA. Northivest Georgia. Dark mulatto soil Metamorphic region. Gray sandy soil . 82 Sandy soil 288 Mulatto soil Central cotton belt. Open pine woods. Hummock soil ... Red hills soil Crowleys Ridges, Greene County. Van Buren do Perry County Jefferson County Arkansas County Independence County . Batesville Phillips County Jackson County ... Saline County Miller County Crowleys Ridge do Moreau Prairie County Lee County - Arkansas County . Union County Hempstead County . Pulaski County. . . Tell County Pope County White County Sebastian County. Marion County Pulaski County . . Benton County.. Madison County. Newton County . Izard County Pulton County. . . Marion County . . Cedartown. Clarksville Douglas County Jonesboro Thomasville 6 1.04 Decatur County 2.37 Stewart County 6 2.71 a Oxid of manganese and iron. In, 10 Per cent. 80.60 89. 90 93.52 71.17 81.24 83.73 91.59 90.40 91.63 90 79.42 91.87 91.79 88.40 92.33 88.40 86.46 92.12 35.14 90.91 90.84 90.31 88.85 83. 24 88.96 87.34 92. 20 91.85 90.85 85.08 77.35 90.80 1.83 2.57 3.44 Per cent. 0.40 3.09 .61 10. 34 6.09 5.29 2.81 3.39 2.19 1.52 4.89 4.91 2.94 5.24 3.46 3.19 3.09 4.38 4.51 Per cent. 1.01 2.36 1.54 7 4.64 3.31 2.12 2.71 2.07 5.27 4.72 1 96 2.29 2.79 2.02 2.79 3.97 1.87 2.54 2.27 2.94 3.05 2.99 6. 94 3.34 'a2.86 4. 64 4. 39 1.19 2.33 3.14 4.79 7.24 3.34 2.56 2.16 2.46 4.49 5.36 a2.67 2.60 3.18 3.05 1.58 1.09 10.60 .85 2.66 2.17 .93 1.13 4.05 Per cent. 0.13 .58 .31 .21 1.01 .71 .44 .21 .30 .30 .21 .53 .21 .19 .29 .12 .16 .24 .03 .07 .13 Per cent. 0.08 .21 .16 .18 .46 .22 .22 .21 .26 .19 .18 .16 . 12 .19 .13 . Id .13 .\Ji .12 .05 .06 .04 .09 28. 13 .06 .02 .21 .02 .18 .06 .10 .05 .21 .07 .14 .17 .14 .12 .04 .03 .19 .07 .08 .11 .19 .14 .17 .24 .12 .11 .04 .29 .04 .02 .30 .13 .23 .08 .01 .05 .24 .05 .07 .22 166 THE COTTON PLANT. Analyses of typical soil* of the cotton States — Continued. Kind of soil. Georgia— continued. Long-leaf pine and wir grass region. Sandy soil Sandy jiino woods Coast region. Live-oak soil LOUISIANA. Alluvial region. Reddish loam Front-lands subsoil Attakapas prairie region . Black prairie soil. Pine prairie soil. . Long-leaf pine region. Pine-kills subsoil Oak uplands. Red -clay soil Red upland subsoil. Yellow sandy soil . . MISSISSI1T1. Northeastern prairie region. 172 Prairie soil. 125 Black prairie, on which cotton rusts badly. 175 I Bald prairii Heavy clay soil Dark-loam hummock soil. Black hummock Flatwoods region. Heavy clay soil White-oak flatwoods . Yellow -loam region or oak uplands. Flatwoods hills Oak upland soil Red-hill soil 228 I Sandy oak uplands 2'6 j Brown-loam table ds.. 156 i Big black hummo soil-. 396 Buckshot soil 376 Light front-lands soil 390 Stiff buckshot soil Cane hills. 237 Loess 117 White hummock soil . 116 Upland loess Central prairie region. Black prairie loam soil . . Black prairie soil Upland prairie soil Bald prairie soil Hog- wallow prairie soil . Hog-wallow upland soil. Locality. Screven County Telfair County Liberty County Bed River bottom Girards New Iberia Serpent Bayou. Vernon County Vienna Mansfield Bastrop hills Monroe County. Buena Vista Booneville Kemper County Tippah County . Lee County Pontotoc County Chickasaw County . . . Lafayette County. . . Greensboro Kosciusko Lafayette County. . . Benton County Sumner County Coahoma County ... Indian Bayou. . . Issaquena County. . \f. © 12 10-20 10 5-12 12 Claiborne County Bayou Pierre Claiborne County. Rankin County. Smith County . . Clarke County.. do Tallahalla Creek Clarke County — 8-10 ..... i er cent. 0.87 1.10 9.90 2.03 16.08 6.58 5.75 3.60 13. 22 5.96 3.80 1.63 2.32 1.64 3.64 10.90 4.04 20. 70 2.56 6.45 4.46 16.10 7.83 Per cent. 1. 10 2.26 1.41 G.37 4.83 1.71 11.71 4.99 4.87 Per Per cent. ; cent. 0.67 | 0.32 .97 .09 .29 1.C9 4.10 2. 78 1.11 3.21 15. 93 8.62 3.55 14,22 11.16 14.37 16. 07 3.43 3.02 10. 30 1.77 1.27 1.87 17. 50 2.65 6.28 3.47 4. 16 2.57 10.54 1.39 2.51 5.51 7. 65 7.09 3.42 1.59 5. 90 3.06 1.63 1.84 10.50 2.53 4.80 2.87 2.80 1.85 5.82 3.27 1 2.95 7. 25 4. 77 10.36 17. 83 I 7. 34 7.45 I 4.09 10. C6 4. 12 6.37 I 3.95 .15 .24 .73 .12 .55 .17 .6] .23 1.10 .51 .13 .36 .57 .69 .59 .53 .36 Per cent. o. 13 .04 CLIMATOLOGY AND SOILS. 167 Analyses of typical soils of the cotton States — Continued. No. Kind of soil. Locality. 03 O 'S o 3 2 73 "3 3 < a O u =H o V, o O Ph 'o 03 o °C o » CD O Jl Ph s 3 205 Mississippi — continued. Long-leaf pine region. Simpson County Lawrence County In. 6 Per cent. 2.02 Per cent. 1.48 Per cent. 1.26 3.65 2.70 .61 .46 .52 .52 40 .30 4.79 1.58 .70 .75 2.82 .60 .74 1.25 1.56 11.70 3.66 4.69 3.86 6.84 4.77 5.29 3.69 9.58 1.03 .61 1.61 Per cent. 0.07 .27 .25 .11 .06 .05 .08 .16 .05 .20 .14 .13 .09 .12 .13 .08 .12 .14 .21 .40 .43 .70 .51 .25 .33 .58 .75 .05 .11 .11 Per cent. 0.07 .10 .10 .01 .02 .10 .11 .30 .06 .14 .12 .02 .06 .05 .21 .04 .10 .09 .13 .06 .32 .24 .31 .06 .34 .32 .36 .09 .17 .19 Per cent. 0.06 292 Red loam subsoil White hummock soil Sandy soil Sandy sea Island cotton soil. Sandy Sea Island cotton subsoil. NOKTH CAROLINA. Seaboard region. Dark mucky soil Cypress and green swamp. Long-leaf pine region. Gray sandy soil Gray sandy loam 8-18 6.69 5.11 6 30 .06 .12 18) 214 88 90 21 25 6 1 7-1 .82 .85 .46 .32 6. 3.33 7.40 2.90 1.48 1.56 3.80 2.85 1.85 3.98 3.67 26.54 5.03 3.10 7.73 10.30 7.77 7.12 6.98 17.30 .28 .91 1.47 .06 Jackson County do Mattamnskeet Lake 12 12 12-20 2.45 .03 3.60 4.02 1.58 3.67 2.95 4.72 2.06 1.48 3. OC 2.85 5.S7 «86. 98 «7G. 50 «77 70 .02 .10 .12 .12 .10 13 11 IS Weldon .12 37 11 12 5-20 12 12 12 5-20 .45 do Oak uplands. .05 2 6 .07 SOUTH CAROLINA. The coast region. .04 R The upper pine belt. Savannah Itiyer Oran geburg County . . . do .11 Red-hills formation. .06 2 .08 11 Mctamorphie region. Spartanburg Gills Station .03 16 TENNESSEE. Upland of vest Tennessee. .24 17 9^ Murfreesboro Maury County Hermitage Dell Meade (Nashville) 12 10-20 7 15 3.97 .14 «, Central basin. .12 8 12 14 10 . doa^:::::::::::::::::: Poplar land soil do 7-15 «82.48 11-23 a81.87 10-22 «84> 03 8 20 n-riS 96 .14 .55 .47 8.38 1 TEXAS. Timbered uplands. Sandy soil Sandy upland soil Dark loamy soil • 12 24 8 .98 3.36 1.01 .27 3 4 Mineola Palestine .03 .15 a Insoluble matter and soluble silica. 168 THE COTTON PLANT. Analyses of typical soils of the cotton States — Continued. No. 16 Kind of soil. TEXAS — continued. Timbered uplands — Cont'd, Red glauconitic soil Sandy prairie soil do Black waxy prairie soil . . Black prairie upland soil. Dark sandy prairie soil . . Central black prairie region, Black sandy prairie soil . . Black waxy prairie soil . . Black waxy soil Northwestern red-loam re- gion. Bed-loam soil Alluvial or river lands. Dark-loam second-bottom soil. Brazos Valley soil Colorado Valley soil Locality. Leo County Tehuacana Pierce's Junction. Victoria Chapel Hill Corsicana Ennis Cleburne McKinney Jacksboro Red River , Granbury Austin cS '3 a k er ton) gave equal financial profit. At the Georgia Station the use of commercial fertilizers was almost always profitable, the percentage of profit ranging from 5 to 250. The stations of Louisiana, Mississippi, ]Sorth Carolina, and South Carolina obtain similar results from experiments, and indicate that " the application of fertilizing material to cotton seems, with few excep- tions, to be profitable." W. B. Dana2 states that in 1878 "the increased productiveness due to the use of commercial fertilizers is estimated to be 50 per cent. The effect does not all pass off the first season, but in about the proportion of 70 per cent the first year, 20 per cent the second year, and 10 per cent the third year." 3 To the teaching of these specific experiments may be added the gen- eral experience of the great bulk of the cotton planters, and it may be accepted as proven that cotton responds favorably to artificial manur- ing, and that upon most of the soils of the cotton States all kinds of manures, including concentrated commercial fertilizers at the prices at which they are commonly held, are profitable when judiciously used. COMPARATIVE VALUES OF COMMERCIAL FERTILIZERS AXD HOME MANURES. Results of experiments on this point also vary considerably with the soil and season. In Alabama, green manuring appears to have been most profitable upon both prairie and sandy soils. Peas and melilotus both gave good results; pea vines appeared to be the best fertilizer for cotton ; peas were more economical for green manuring for one season, 'For convenience, reference is omitted to the specific title and page of authorities consulted in the preparation of this article. A short bibliography of the subject is given at the end of the article which embraces all the authorities to Avhich reference is made or from which quotations are taken. 2Cotton from Seed to Loom. 'Hammond estimates that in 1889 627,899 tons of fertilizers were used on cotton in the United States, and that the use of this amount of fertilizer resulted in an increased crop of 728,337 bales of seed cotton, or a little over 9 per cent of the crop of that year. 182 THE COTTON PLANT. inelilotus for two; stable manure generally gave good results, lasting in effects; upon canebrake soils, both drained and undrained, crushed cotton seed and stable manure each gave small returns, commercial fertilizers none. Upon a field with sandy soil which had not been culti- vated for many years stable manure, contrasted with chemical manures of various kinds and in various proportions, produced the largest increase and the largest profit per acre, but it was noted that the amount applied was at the rate of nearly 2 tons per acre, or one-half ton more than the amount annually saved from each mule kept. "There is no question about the efficacy of good stable manure properly used, but the available supply is too small." In Arkansas, cotton seed and cotton-seed meal gave best results when tested against acid phosphate and kainit separately. On worn sandy bottom lands almost continuously planted in cotton for thirty years cotton-seed meal and stable manure each gave better results than chemical manures, and better results when used alone than when mixed with acid phosphate and kainit. "There is no better fertilizer for cotton than stable and barnyard manure." Other experiments indicated, how- ever, that stable manure (from feeding cotton seed and pea-vine hay) extended the growing season of the plant, delayed maturity of the crop, and hence decreased the possible yield and profit. In Georgia, cotton seed and stable manure alone were found unprofit- able as compared with the same composted with acid phosphate, and gave less profit, when used in amounts of equal cost, than chemical manures. In Louisiana, cotton seed and stable manure alone were of doubtful profit as compared with chemical manures. "Manure from the farm should be reenforced with cotton-seed meal and composted with acid phosphate. * * * The compost is the best manure in the world for cotton." The formula recommended for the compost is — Green cotton seed bushels.. 100 Stable manure . . . .. do 100 Acid phosphate pounds.. 2, 000 Almost as effective as the compost was a homemade chemical manure, constructed as follows: \ Pounds. Acid phosphate '. 1, 100 Cotton-seed meal 700 Kainit 200 In Mississippi, commercial fertilizers were more profitable than stable manure or cotton seed alone, but paid best in connection with an abun- dance of organic matter. Composts variously proportioned gave best results. In l^orth Carolina, barnyard manure was found to be especially effect- ive, partly on account of its after effects, and somewhat the best of all fertilizers. Its first cost ($1 per load), however, detracted from the profit, and a combination with acid phosphate was much more profitable. THE MANURING OF COTTON. 183 Home composts gave generally good results, and next to these a home mixture of — Pounds per acre. Acid phosphate 200 Cotton-seed meal 100 Kainit 50 KIND OF FERTILIZER (CHEMICAL MANURE) REQUIRED BY OR BEST SUITED TO COTTON. Assuming phosphoric acid, potash, and nitrogen in suitable compounds to be the three chemical substances proper and possible to be used in the fertilization of cotton, the experiments have been mainly conducted with a view to determine the relative importance of these, the best form of each (i. e., of the compounds available in commerce), and the propor- tions of each in a mixed fertilizer most suitable to the requirements of the cotton crop, regard being had to the character of the soil to which they were applied and account being taken of the profit afforded. In Alabama, in 188S, experiments were made upon a sandy drift soil to determine the proper ratio of nitrogen to phosphoric acid in fertilizers for cotton . The amount of phosphoric acid was constant — 200 pounds of English superphosphate (12 per cent soluble) per acre — and the amount of nitrogen (in dried blood and cotton-seed meal) varied so as to furnish 1 part nitrogen to 1, 2, 4, 6, and 8 parts phosphoric acid. The smallest quantity of nitrogen employed gave as good results as larger quantities. No difference was observed in the two sources of nitrogen. In 1889 coop- erative experiments under direction of the College Station were made on 9 farms furnishing typical soils of the State. The fertilizers used and the amounts per acre were: For nitrogen, sulphate of ammonia, 80 pounds; nitrate of soda, 100 pounds; cotton-seed meal, 200 pounds; for phosphoric acid, dissolved boneblack, 200 pounds; for potash, kainit, 100 pounds. Green cotton seed (960 pounds per acre) and stable manure (3,000 pounds per acre) alone and in combination with acid phosphate were also used. The fertilizers were applied singly and in various com- binations to fifteenth-acre plats without duplication. Some of the experi- mentsprovedtobeof little value, owing to mistakes and omissions; others indicated with some clearness that x»hosphoric acid was the ingredient chiefly needed in the soils tested — sandy and brown loam, with clay sub- soil. An experiment was also made with cotton on newly cleared land, in which acid phosphate was applied on two plats, acid phosphate and cotton- seed meal on two, and no manure on one. The results indicated that the natural soil did not furnish sufficient nitrogen and was very deficient in phosphoric acid for the requirements of the crop. In 1890 an experiment was made on fifteen plats in a field which had not been cul- tivated for many years. The fertilizers used were sulphate of ammonia, dissolved boneblack, and kainit, singly, two and two, and all three together. Floats, alone and in combination, separately, with sulphate of ammonia and green cotton seed, was also used, as also stable manure and green cotton seed singly. Contrast was made with plats receiving no manure. The results indicated that this soil needed nitrogen and 184 THE COTTON PLANT. potash, but was most deficient in phosphoric acid for the production of the crop. This experiment and another on a sandy drift land long in cultivation indicated that floats, in connection with cotton seed, was more profitable than acid phosphate. In 1S91 cooperative experiments were made on 30 farms in various parts of the State with different fertilizers in different amounts and combinations. The experiments were not perfectly accurate, but indicated certain conclusions. Potash did not seem to pay; phosphates applied alone did not have much effect; nitrogenous fertilizers in all forms gave an increased yield. In 1893 certain experiments indicated that nitrogenous fertilizers (cotton-seed meal and nitrate of soda) alone on cotton pay on sandy lands, pro- viding there are good rains following their application. The general indications afforded by the great number and variety of cooperative experiments made since 1888 under the auspices of the Alabama Station, upon a variety of soils of the State, the majority of which were sandy, are that a complete fertilizer is needed for cotton. Phosphoric acid is often the controlling element, and a sufficiency of nitrogen is frequently lacking in the soil. Potash alone does not pay. Phosphates applied alone have some effect, but much less than when combined with nitrogen. Nitrogen, particularly in organic forms, is profitable, especially in connection with phosphates. The unfertilized soil of the station needs nitrogen, potash, and imosphoric acid. It is especially deficient in the latter. " In new ground the decomposition of the vegetable matter in the soil did not furnish all of the nitrogen needed by the cotton ; the increase from phosphates alone was satis- factory, but the increase caused by the addition of nitrogen did not justify its use." As to floats, the experience of several years indicated that a part of the phosphoric acid becomes available to the plant the first season, but the solubility is much facilitated by combining the floats with cotton seed or cotton-seed meal. In Arkansas, in 1889, experiments were made on sandy bottom land which had been almost continuously planted in cotton, without manur- ing, for thirty years. Acid phosphate, cotton-seed meal, and kainit were used singly and in combination ; also stable manure and composts in different amounts. Nitrogenous manures alone were profitable. Neither acid phosphate nor kainit alone paid. All of the different plats on which cotton-seed meal was used, either singly or in combina- tion, gave some profit, and u this was due not to the acid phosphate or kainit, but to the cotton-seed meal." These results were confirmed by similar experiments made in 1891. Cooperative experiments made in 1888 at five points in the State, and repeated in subsequent years, indi- cated that a complete chemical fertilizer is needed for cotton. A com- bination is provisionally recommended of — Pounds per acre. Acid phosphate 200 Cottou-seed meal 200 Muriate of potash 50 THE MANURING OF COTTON. 185 In Georgia a series of excellently arranged and very carefully con- ducted experiments have been in progress upon tlie station farm since 1890. The soil of the station is somewhat irregular in character, but is, for the most part, a gray sandy loam underlaid by yellow clay and, previous to the institution of the experiments, had been in continu- ous cultivation for a number of years. Fertilizing materials in great variety and in many different combinations were used. The results of the experiments have not been strictly accordant, but the folio wing gen- eral conclusions seem to be provisionally warranted. Cotton requires a complete manure, i.e., one containing soluble phosphoric acid, potash, and nitrogen. Neither phosphoric acid nor potash give as good results alone as when combined with each other. Phosphoric acid alone largely surpasses no manure. Potash alone is doubtful; sometimes it affects the yield injuriously. Nitrogen alone has little or no effect, but has very decided effects when mixed with phosphoric acid and potash. In some cases nitrogen seems to be the controlling element in a ferti- lizer, but, on the whole, phosphoric acid is most effective in increasing the yield. Cotton-seed meal (and cotton seed) and nitrate of soda seem to be the best forms of nitrogen for cotton and are about equal in value, proportionately to the content of nitrogen. There is little or no difference in the value of kainit and muriate of potash. The phos- phoric acid in floats and Florida soft phosphate is not in a sufficiently soluble and available condition to answer the needs of the cotton crop. The best proportions of the three elements in a complete fertilizer for cotton are, approximately, nitrogen, 1 part; potash, 1 part; phosphoric acid, 3J parts. Of such a complete fertilizer the quantity to be used per acre should be an amount furnishing nitrogen, 20 pounds; potash, 20 pounds; phosphoric acid, 70 pounds. In Louisiana admirably conceived and carefully conducted series of field experiments in the fertilization of cotton have been made both at the State Station, at Baton Eouge, and at the North Louisiana Station, at Calhoun, beginning in 1886 and still in progress. Plats of uniform size were manured with nitrogenous, phosphatic, and potash fertilizers of different hinds aud in different proportions, separately and in great variety of combination. The questions tested were: (1) Do these soils (the worn "sandy lands" aud ared lands" of Louisiana) need nitrogen to grow cotton profitably? If so, in what form can it be best presented, and in what quantities per acre? (2) Do these soils need phosphoric acid? If so, which is the best form, and in what quantities per acre? (3) Do these soils need potash? If so, which is the best form, and in what quantities per acre? The results of the experiments were, in some instances, inconclusive, and in some apparently contradictory, as the seasons and the condi- tions varied. On the whole, however, the followiug conclusions seem justified as the result of the entire series of experiments : (1) These soils need nitrogen, and nitrogenous manures may profit- ably be used in the fertilization of cotton. 186 THE COTTON PLANT. (la) All forms of nitrogenous matters (vegetable, animal, and min- eral) are satisfactory and profitable, but, on the whole, they stand in the following order of excellence: (a) vegetable (cotton seed and cotton- seed meal); (b) animal (dried blood, fish scrap etc.); and (c) mineral (sulphate of ammonia and nitrate of soda). (lb) One ration (24 pounds) of nitrogen per acre is more profitable than larger quantities. (2) These soils need phosphoric add. Phosphatic manures may be profitably used in the fertilization of cotton. They are not so neces- sary (upon these soils), however, as nitrogen. (2a) The soluble forms of phosphoric acid (in dissolved boneblack and acid phosphate) are emphatically better than the insoluble forms (in floats and similar materials). (2 b) One ration (24 pounds per acre) of phosphoric acid is more profit- able than larger quantities. (3) Potash in no form, either alone or combined with other manures, is needed for these soils. Potash manures are not profitable in the fertilization of cotton. "It is very certaiu that phosphoric acid is needed to grow cotton successfully, but in small quantities and combined always with nitrog- enous manures." In Mississippi experiments in the fertilization of cotton were made at the College Station and at Holly Springs in 1888-1893. Tbe results indicate that on upland soils the fertilizer should be rich in organic matter and nitrogen and contain more potash than phosphoric acid. On sandy valley lands the phosphoric acid should predominate. Lime soils require large quantities of potash. On soils poor in lime potash was not needed or did not pay. On black prairie lands the value of concentrated fertilizers was not definitely indicated. The results in different years were conflicting. Cotton-hull ashes were found to be an excellent form of potash. In North Carolina experiments "on representative soils of the chief geological areas in the State" were conducted in 1890-1894. Stable manure gave best general results in yield, but was not always most profitable on account of initial cost. Next to stable manure, a "com- plete fertilizer" gave best results, and the proportions per acre recom- mended are: Pounds. Acid phosphate 200 Cotton-seed meal 100 Kainit 50 Acid phosphate alone was, for the most part, profitable. Cotton-seed meal alone was profitable in the majority of cases. Kainit alone was unprofitable except in the case of the poor sandy lands of eastern North Carolina. In South Carolina a very elaborate and most carefully conducted series of experiments was made upon the station farms (two), situated THE MANURING OF COTTON. 187 in different sections of the State, and extending over three years— 1888-1890. The soils selected were typical of the "upland" soils of the cotton States, and were "very thin, being greatly exhausted by years of improvident culture." Applications of fertilizers (phosphatic, nitrog- enous, and potash) of various kinds were made, separately and in various combinations, and in different amounts, but more particularly in the approximate quantities and proportions shown by existing analyses of the cotton plant to be necessary for the requirements of the crop. The details of the experiments and the results have been reported in a bulletin of the United States Department of Agriculture.1 The conclusions reached are in part as follows: (1) Cotton requires nitrogen, phosphoric acid, and potash. (2) Of the three, phosphoric acid is relatively the most important and controls the action of the other two. It can be used alone with some advantage to the crop, but much more effectively in connection with potash and nitrogen. (3) Nitrogen is relatively more important than potash. It can only be advantageously used in combination with phosphoric acid or phos- phoric acid and potash. (4) Potash, like nitrogen, is of little value to cotton when applied separately; it must be combined with the other constituents. (5) Expressed in terms of the latest analyses of the cotton plant, the proportions and amounts of nitrogen, phosphoric acid, and potash required are as follows: Between three-sevenths and four-sevenths nitrogen, about four and one-fourth phosphoric acid, and between one- third and one-half potash. With proper allowance for the cost, as well as the effect of each application, the requirements maybe more exactly given as follows: Nitrogen, 0.13; phosphoric acid, 1.16; potash, 0.38. In other words, the required proportions are: Nitrogen, 1; phosphoric acid, 2^; potash, f ; and the amounts called for by a crop yielding 300 pounds of lint per acre are: Nitrogen, 20 pounds; phosphoric acid, 50 pounds; potash, 15 pouuds. (6) The amount of phosphoric acid determines the amount of nitro- gen and potash. With a given amount of the first, only certain amounts of the last two can be profitably used. (7) Potash can be as effectively supplied by muriate of potash or kainit as it can by sulphate of potash. (8) Phosphoric acid is of value to cotton in proportion to its solu- bility; hence, the several kinds of phosphatic manures can not be indifferently employed. Preference must be given to acid phosphates containing considerable percentages of soluble phosphoric acid. Insol- uble phosphoric acid in slag, floats, or marl is of little direct value to the crop upon which it is applied, and even granting that its effects in the soil may be lasting they are not, in the long run, sufficiently pro- nounced to meet the interest on the capital invested in the application. 'Farmers' Bui. 14. 188 THE COTTON PLANT. (9) Inorganic, organic, and mixed nitrogen are of very nearly equal value to cotton. The slight difference is in favor of the last two. Stable manure containing organic nitrogen is the best fertilizer of its class, and is lasting or cumulative in its effects. The organic nitrogen of stable manure, to the amount of 50 per cent, can be fully replaced by the inorganic nitrogen of nitrate of soda. As between cotton seed and cotton-seed meal, there is a slight difference in favor of the latter. Whole cotton seed is as efficacious as ground cotton seed. Inorganic nitrogen in nitrate of soda is about as valuable to cotton as organic nitrogen in cotton seed or cotton-seed meal. The results obtained in Georgia and South Carolina are worthy of special consideration in this connection, as the experiments yielding them were conducted specifically for determination of the points now under discussion. THE AMOUNT OF FERTILIZER PER ACRE GIVING BEST RESULTS. The experiments bearing upon this question are somewhat meager and the results uncertain. The amount of fertilizer which may be judiciously and profitably employed is shown clearly to depend upon the character, condition, and previous treatment of the soil, and to some extent upon the season. Very few systematic experiments have been made to test this specific question. In Alabama one series of experiments indicated that an application of 1,000 pounds per acre of a complete fertilizer was not as profitable as one of 500 pounds, although the yield was somewhat increased. In Georgia large doses of fertilizer applied at planting or during the earlier periods of growth resulted in earlier maturity of the crop, with- out, however, sensible increase in profit. The results of experiments conducted for several years on series of plats of gravelly gray soil with yellow subsoil, in which fertilizers were applied at the rates of 400, 600, and 1,200 pounds per acre, indicated — (1) That, while heavy doses of fertilizers do not give a corresponding increase in the yield of cotton, or so large a percentage of profit, yet such heavy applications, within reasonable limits, are judicious, pro- vided the land is in good condition. (2) That the limit or maximum amount of fertilizers that can be safely and profitably applied to land in good condition varies consider- ably, say from 500 to 1,000 pounds per acre, according to seasons, vari- ety of cotton, etc. In these experiments the maximum amount that was immediately profitable was probably between 500 and 700 pounds per acre. It is concluded that, in general, the most effective amount of ferti- lizer was 652.6 pounds per acre, compounded as follows : Pounds. Acid phosphate ,468 Nitrate of soda 130 Muriate of potash 54. 6 652.6 THE MANURING OF COTTON. 189 or such an equivalent amount of similar mixtures as will furnish per acre, approximately, phosphoric acid, 70 pounds ; nitrogen, 20 pounds ; potash, 20 pounds. It has been shown that $8 worth of a well-balanced fertilizer may be expected to increase the yield of seed cotton on 1 acre 1,000 pounds. But such results can only be attained by concentrating the fertilizer on the best land, not by scattering it at the rate of 100 or 200 pounds to the acre over a large, worn-out plantation. The mistake should not be made of applying large amounts of concentrated fertilizers on thin, worn-out land. The larger the application the more important is it that the land be in the best possible condition. In North Carolina heavy applications of stable manure, while some- what proportionately increasing the yield, were not profitable. In South Carolina it is concluded that "the amount of phosphoric acid and proportionate amounts of nitrogen and potash can not be indefinitely increased with the expectation of obtaining a correspond- ing increase in the crop. The gain in crop does not keep pace with increase of fertilizers, and a point is speedily reached beyond which this gain is not sufficient to meet the additional cost of the heavier applications. The soil can not be profitably forced ; the application of fertilizers must be regulated by its mechanical as well as chemical con- dition." The maximum quantity of fertilizer that can in general be used with advantage is concluded to be an amount that will furnish per acre phosphoric acid, 50 pounds; potash, 15 pounds; nitrogen, 20 pounds. BEST MODE OF APPLICATION OE FERTILIZERS TO COTTON. In Alabama experiments in 1887 indicated that broadcasting compost and stable manure gave better results than application in the drill. In Georgia the results of general experiments indicate that " it is by no means necessary, nor is it desirable, to broadcast the fertilizer when less than 1,500 pounds are to be applied to an acre of corn or cotton or other wide-row crop. If only 500 pounds are to be applied, distribute it in a deep furrow and mix it by running two scooter furrows through it. If more than 500 pounds, then divide the amount between the center furrow and the two listing furrows. Broadcast manuring should, as a rule, be confined to crops that are planted broadcast, as small grain, grass, etc." The experience and the practice at the stations generally substan- tiate the conclusion reached in South Carolina that " fertilizers may be iudifferently drilled or broadcasted where they are liberally applied, but drilling is to be preferred where small amounts are employed." BEST TIME OF APPLICATION OF FERTILIZERS TO COTTON. A number of experiments have been made to test the effects of inter- cultural applications of fertilizers, the results ®f which, however, are for the most part discordant and inconclusive. .190 THE COTTON PLANT. Iu Alabama one set of experiments, in 1888, indicated that one application of the fertilizer in the drill before planting gave best results. [Another, in 1890, was very carefully conducted] in order to teat the efficacy of the application of additional fertilizer during the growth of the plant in prolonging its fruiting period and increasing the yield. Two hundred pounds of cotton-seed meal per acre were applied at the second plowing of the cotton, June 18, and cov- ered lightly with the scrape. Two hundred more were applied in the same way at the last plowing, July 30. These were applied to two plats to which 200 pounds of cotton-seed meal and acid phosphate, mixed in equal parts of each, were applied in the drill before planting, and were compared with a third plat to which the same quantity of cotton-seed meal and acid phosphate were applied at planting, but to which no subsequent applications were made. The average increase caused by the additional applications was 339 pounds of seed cotton per acre. The iutercultural applications had the effect of continuing the growth and fruitfulness of the cotton after that on plat 3 had ceased to grow. In 1893, however, it was found that 200 pounds of mixed nitrate of soda and cotton-seed meal applied (to previously fertilized plats) in June was as profitable as 100 pounds in June and 100 pounds in July. The addition of cotton-seed meal as late as August 13 was not profitable. Tn Georgia it has been found that marked effects result from inter- cultural fertilization, or successive applications of fertilizers during the growing season. When a heavy application of a readily available fertilizer is to be made, it would be advisable to divide it into at least two doses, and possibly more. In Louisiana the conclusion is reached that fertilizers for cotton should all be applied at time of planting. A second and third appli- cation is not profitable. MISCELLANEOUS EXPERIMENTS. Incidentally to the main objects for which the experiments in the fertilization of cotton were instituted, certain indications on miscellane- ous and minor points have been afforded. The general experience at the stations and elsewhere is to the effect that chemical manures generally, and especially nitrogenous and phos- phatic manures, hasten the maturity of the crop. Stable manure in some instances (Arkansas, North Carolina) delayed maturity beyond the fruiting period. The cumulative effect of manures in the soil is fairly well evidenced in several cases. Nitrogenous manures increased the yield the second season without additional fertilization (Alabama, Arkansas), but not the third season (Alabama). Phosphatic manures increased the yield, without additional fertilization, the second and third seasons (Ala- bama). The cumulative effects of heavy applications of a complete fertilizer were manifest the second and the third years (Georgia). Floats alone gave a greater increase over no manure the third year after application than in the first or second year (Alabama). Kainit is recommended as a specific for rust and blight in cotton, to THE MANURING OF COTTON. 191 be used in connection with cotton seed or cotton-seed meal (North Carolina). Kainit appears to retard the appearance of blight (Ala- bama). Kainit retards the opening of the bolls (Arkansas). "Marl, alone or in combination with commercial fertilizers, is of no direct value to cotton. Applied upon leguminous crops, which are to be turned under as a preparation for cotton, its indirect value is great" (South Carolina). Air-slacked lime mixed in the drill with acid phos- phate and floats had no apparent effect upon the crop (Alabama). Applications of copperas are without effect upon cotton (feouth Carolina). Nitrate of soda should generally be applied with the other fertilizers at the time of plantiug (South Carolina); but, on the other hand, it may- be profitably divided into two applications, the second not to be later tban Junel (Georgia). The quantity of nitrogen in the fertilizer seems not to affect the rela- tions between the weight of seed and lint (Alabama). Shallow applications of fertilizers (i. e., at depth of 2 or 3 inches) give better results than deeper applications (Louisiana). There is no advantage in separating the ingredients of the fertilizer and applying them at different depths (Louisiana). It is highly important that the fertilizer be well mixed with a consid- erable portion of soil (Georgia). The cowpea is an excellent green manuring crop in preparation of land for cotton (Alabama, Arkansas, Louisiana, Georgia). The most profit- able method of application is to gather the peas, or cut the vines for hay, and turn under the stubble with addition of the manure from stock fed with the hay (Alabama, Arkansas, Georgia). GENERAL CONCLUSIONS. In reviewing the results of the experiments conducted at or under the auspices of the experiment stations, and taking into account the general experience of successful cotton growers, certain general con- clusions on the subject of the fertilization of cotton may be accepted as tentatively established : (1) Cotton is a plant which responds promptly, liberally, and profit- ably to judicious fertilization. (2) By judicious fertilization the maturity of the crop may be has- tened and the period of growth, from germination to fruiting, so short- ened as to materially increase the climatic area within which cotton may be profitably grown. (3) As is the case with most other crops, the profit from manuring cotton with concentrated fertilizers is much enhanced by antecedent proper preparation of the soil. It pays to bring cotton lands up to a condition of good "tilth" by mechanical treatment, and especially by incorporating in them liberal quantities of organic matter. Upon lands 192 THE COTTON PLANT. in such condition fertilizers of all kinds yield more profit, either from small or large applications, than upon lands not so treated. (4) Benovatiug crops, and especially the cowpea, furnish an efficient and economical method of bringing cotton lands into condition to respond most liberally and profitably to tlie application of concentrated manures under cotton. The most profitable plan of employing the cow- pea for this purpose on cotton is to gather the peas at maturity, cut the vines for hay, and turn under the stubble along with the manure result- ing from feeding the hay to stock and cattle. (5) Barnyard manure and similar bulky manures are more efficient and profitable as soil renovators than as specific fertilizers for cotton. They should be broadcast liberally and used rather as soil improvers than as immediate fertilizers. The same is probably true of cotton seed, except where the price to be had for the seed at cotton-oil mills justifies the exchange of seed for cotton-seed meal, to be used as the source of nitrogen in a concentrated manure. If, however, only small quantities of such manures are to be had, and it is desired to use them as direct fertilizers, it is more profitable to compost them with acid phosphate (preferably containing a small percentage of potash) than to use them alone. It is more profitable to compost directly in the drill at time of planting than in heaps previously. (6) Cotton may wisely be assigned a place in a judicious rotation sys- tem. Upon lands devoted to staple crops a three years' rotation — small grain, corn (with peas), cotton — is judicious. Each crop in the rotation should be appropriately fertilized. It is in evidence that the cumula- tive effects of such manuring upon the succeeding crop are marked. (7) Upon the great majority of the soils of the cotton-growing States it is advisable and profitable to use, as a concentrated fertilizer, a "complete manure," i. e., one containing soluble phosphoric acid, avail- able potash, and available nitrogen, rather than a manure containing only one or two of these iugredients. Nitrogen, however, may proba- bly be advantageously omitted from the concentrated fertilizer, in whole or in part, when the soil has previously been liberally supplied with this ingredient, through barnyard manure, green manuring, etc. (8) "Soluble" phosphates are very much to be preferred in the fer- tilizer for cotton to those which are not soluble. (9) There is no great difference, if any, in the agricultural value and profit, when used in the fertilizer for cotton, of the various soluble pot- ash salts to be had in commerce, except proportionately to the price and content of actual potash. (10) Of the nitrogen compounds available for use in fertilizers the organic forms (vegetable and animal) are perhaps best suited to cotton, if one form alone be used, although nitrate of soda is probably nearly if not quite of equal value. Further experiments are needed to deter- mine the efficacy of mixing various nitrogen compounds in different proportions. THE MANURING OF COTTON. 193 (11) The most judicious proportions of soluble phosphoric acid, pot- ash, and nitrogen in a complete fertilizer for cotton can not be said to have been as yet determined with entire accuracy. Those suggested by Georgia — nitrogen 1, potash 1, phosphoric acid 34 — and by South Carolina — nitrogen 1, potash f, phosphoric acid 2^ — perhaps approxi- mate reasonable accuracy. In the light of present information, perhaps nitrogen 1, potash 1, phosphoric acid 2§ or 3 would not be injudicious proportions for general use. (12) The amount of concentrated fertilizer which may profitably be used per acre varies widely with the nature and condition of the soil, the seasons, and other circumstances. For an average soil in fairly good condition perhaps the maximum amounts indicated by Georgia (nitrogen, 20 pounds; potash, 20 pounds; phosphoric acid, 70 pounds), or by South Carolina (nitrogen, 20 pounds; potash, 15 pounds; phos- phoric acid, 50 pounds), or an approximate mean of the two would be the maximum limit of profitable application. The actual weight of the complete fertilizer furnishing these quantities would, of coarse, vary with the percentage composition in nitrogen, potash, and phosphoric acid of the materials used to make the fertilizer. If the commercial "ammoniated" fertilizer or other concentrated manure intended for use under cotton should be compounded (as it might very well be, and in some cases is) to analyze approximately — Per cent. Soluble (available) pbospboric acid 9 Potasb 3 Nitrogen 3 then 700 jKrands per acre of such a fertilizer would be approximately the maximum amount that could judiciously and profitably be used, under ordinary circumstances, upon soil in good condition. (13) The concentrated fertilizer should be applied in the drill (not broadcast) at a depth of not more than 3 inches, and well mixed with the soil. (11) All things considered, it is perhaps best in most cases to apply all the concentrated fertilizer in one application at the time of plant- ing. With lands in superior condition, however, or where large quan- tities of fertilizers are used, it is probably profitable to apply half at planting and half at the second plowing. These conclusions, it is believed, are justified by the present state of knowledge on the subject of the fertilization of cotton. They may be accepted provisionally and until modified and corrected by the results of further investigations and experiments, such as are now in progress at several of the experiment stations in the Southern States. In view of the importance of the subject and the unsatisfactory state of knowl- edge concerning it, the writer ventures to suggest that it would prob- ably be wise for some one of the stations of the cotton States to devote a large part of its time and resources to an extensive, thorough, and intimate study of the nutrition, growth, and development of the cotton 1993— No. 33 13 194 THE COTTON PLANT. plant. It is perhaps not hazardous to conjecture that the results of such study might modify materially the apparent conclusions thus far reached on the subject of the fertilization of cotton. METHODS OF MANURING COTTON AT PRESENT IN GENERAL USE IN THE UNITED STATES. The method of fertilizing for cotton at present employed by t\ie Southern cotton grower varies somewhat with differences in soil, climate, capital, etc. The greatest variation, perhaps, is in the preliminary preparation of the land. Some cotton farmers practice green manur- ing, rotation, composting, etc., with regularity, others irregularly, others not at all. There is much greater uniformity observed in the use of concentrated fertilizers, although here, again, there are wide differences in usage, particularly as to the amount of fertilizer employed. From the time of their introduction until the present, the method of applying chemical manures to cotton has been essentially uniform and the same, viz, in the drill. They are very rarely broad- casted. Neither, indeed, as a rule, are composts, stable manure, or cotton seed. A shallow furrow, varying in depth from 3 to C inches — much more frequently 3 than 6 — is opened with the plow and the manure applied by hand (generally through a tin tube, known as a "guano horn," 3 feet long and 2 inches in diameter) or by a mechanical "distributer," much like a grain drill planter and capable of being set to deliver fixed and uniform quantities. The manure is then " listed" on, i. e., covered with a thin layer of soil to the depth of 1 to 3 inches. The seed are dropped upon this, either by hand or from a "planter," and covered with soil to a depth usually of 3 inches. The seed are frequently rolled in ashes, or sometimes in acid phosphate or other fertilizer, before planting. The amount of fertilizer used per acre varies greatly. From as little as 50 pounds to as much as 1,000 pounds per acre is used. The average amount used by the very great majority of cotton growers is probably between 175 and 200 pounds per acre. The fertilizer is for the most part the commercial "ammoniated" arti- cle, although considerable quantities of acid phosphate (with or with- out potash) and home mixtures of chemicals are so used. In the case of composts, such as that prepared by the " Furman formula," for example, the amount used is usually about 400 pounds per acre. The commercial ammoniated fertilizer sold in the Southern States at the present time will average in composition, approximately — Per cent. Soluble (available) pkospboric acid 9 Nitrogen 2 Potash 2 Acid phosphates range in content from 12 to 15 per cent of available phosphoric acid, and are often given a small content of potash, ranging from one-half of 1 per cent to 2 per cent. THE MANURING OF COTTON. 195 The home mixture of chemicals is usually constructed on the formula approximately — Pounds. Acid phosphate 1, 200 Cotton-seed meal 600 Kainit 200 2, 000 MANURING OF COTTON IN OTHER COUNTRIES. Elsewhere than in the United States the culture of cotton is mainly confined to the rich alluvial lands, and a large proportion of the crop is grown without any manuring whatever ; still in most countries some use is made of farm manures. Even on the fertile alluvial soils of Egypt, which are so abundantly supplied with fertilizing materials by the overflowing of the Kile, barnyard manure is applied to the extent of 10 to 15 tons per acre, and " generally speaking, except on the richest land, it is acknowledged by experienced growers that the crop repays the cost of application." In spite of the fact that clover is very generally grown as a prepara- tory crop for cotton in this country, nitrogenous manures as a rule are the most profitable, because the nitrogen of the soil is exhausted by the large crops of cotton and sugar cane which are grown and which return nothing to the soil, and is also dissipated by the rapid nitrifica- tion which goes on under the peculiar climatic conditions of Egypt. "The time at which the manure is applied varies considerably. Some spread it over the land and plow it in before making the ridges. Others ridge the land and spread it in the furrows, subsequently covering it by splitting the old ridges. Either of these methods is suitable and preferable to the system of applying it after planting, which is perhaps more common than the other. When the cotton is a few inches above the ground, the manure is either spread in the furrows and hand hoed in, or a handful is put under the roots of the young plants. This latter method involves more labor than any other, and has no advantages. Fresh manure is not thought so good as that which has been in the heap for two years, and old manure is always used by the best growers. "For the production of manure, earth is used as litter, and the com- position of the resulting manure depends, therefore, to a considerable extent on that of the earth used. It contains but little water, 5 or 6 per cent being an average. As the result of several analyses made by Dr. Mackenzie, the manure may be said to contain : Nitrogen 0.1 per cent, phosphoric acid 0.25 per cent, and potash 1.5 per cent. "Following clover, as cotton almost invariably does, it finds except on very poor laud a sufficiency of nitrogen, if the fodder crop has been grazed. If cut and removed, the case maybe different. After a fallow the land is generally manured, as the land selected for this purpose is of poor quality, and more benefited by its application. Ko artificial fertilizers are applied in practice, and as yet no experiments of a relia- ble nature have been made to ascertain their effect." l 1 Foaden, MS. article on cotton culture in Egypt in the possession of this Office. 196 THE COTTON PLANT. BIBLIOGRAPHY. Tropisclie Agrikultur, Sender, 1857. The Culture of Cotton, J. W. Mullet, 1860. The Cultivation of Cotton, John Gibbs, 1862. The Cotton Question, W. J. Barbee, 1866. Cotton Culture in 1866, N. B. Cloud, U. S. Dept. Agr. Rpt. 1866. System of Farming, David Dickson, 1869. Cotton from Seed to Loom, W. B. Dana, 1878. Encyclopaedia Britannica, 9th ed., Art. Cotton. American Cyclopedia, Art. Cotton. Soils of Cotton States, E. W. Hilgard, 10th U. S. Census, Vols. V, VI. Climatology of the Cotton Plant, P. H. Mell, U. S. Weather Bureau Bulletin No. 8, 1893. Farmers' Bulletin No. 14, U. S. Dept. Agr., 1894. Numerous agricultural journals published in the Southern States, 1845-1895. Agricultural experiment station bulletins. Alabama College Sta. Buls. 3, 4, 5, 12, 22, 23, 33, 34, 40, 42, 52, 57. Alabama Cauebrake Sta. Buls. 4, 7, 11, 14 ; Rpt. 1890. Arkansas Sta. Buls. 1, 18, 23; Rpts. 1888, 1889, 1890. Florida Sta. Buls. 8, 12. Georgia Sta, Buls. 10, 11, 16, 20, 27. Louisiana Sta. Buls. 2, 7, 8, 13, 16, 21, 22, 26, 27, 29, 31. Mississippi Sta. Buls. 24, 29; Tech. Bui. 1 ; Rpts. 1888, 1889, 1890, 1891, 1893. North Carolina Sta. Buls. 65, 71, 89; Rpts. 1881, 1882, 1885. South Carolina Sta. Bui. 2; Rpt. 1889. Tennessee Sta. Bui., Vol. IV, No. 5. CULTIVATED VARIETIES OF COTTON. By S. M. Tracy, M. S., Director of the Mississippi Agricultural Experiment Station. INTRODUCTION. The word "variety" as used in this chapter refers exclusively to the various forms and kinds which are called "varieties" by cotton plant- ers, and is not restricted to the more marked and permanent types which are recognized by the botanists. Of botanical varieties there are but few, while of agricultural varieties there are an almost infinite number, and the names under which the agricultural varieties are known are many times greater than the number of recognizable forms. Cotton is a plant which sports easily, which responds quickly to any differences in environment, soil, climate, treatment, and fertilizers, and which can be greatly modified in form and habit in a very few succes- sive crops. The flowers are large and open, x^ollen is produced in great excess and is readily scattered by the lightest breeze, the stigmas are well above the anthers, so cross fertilization is not only common but usual. Seed from the earlier maturing bolls will produce plants yield- ing a longer lint than will the seed from the later ripening bolls from the same plants. Many varieties which will produce a long, strong, and silky fiber when planted in rich river bottom soil soon lose their superior qualities when grown on the drier and poorer hill lands. A variety which has been grown for some years in the northern belt of the cotton region will mature its entire crop at nearly the same time, while if the same variety be grown for a few years in the southern cot- ton belt the crop will continue to mature through several weeks, though the total yield may be but little greater. With its natural tendency to vary, and with all these forces to stimulate the plant to a change of form and habit, it is easy to see how varieties maybe multiplied indefi- nitely, even without deliberate action on the part of the cultivator. In the preparation of this work an effort has been made to trace the origin and history of all the varieties which have been mentioned in the station publications, together with such others as are known to be common in the cotton-growing region of the United States or to possess special value. In many cases the records of certain of these varieties have been very defective, and it has been impossible to secure accurate data from the originators themselves, so that some varieties of greater or less local prominence are not mentioned- here. We have found it exceedingly difficult to secure exact data in regard to both the percentage 197 198 THE COTTON PLANT. of lint and the length of staple from different varieties, as in many cases the amount of lint has been estimated from the gin yields, which may vary greatly. In experiments at the Alabama Station,1 the averages of eight gin yields varied 1.99 per cent, and in one case as much as 5.25 per cent, when two gins of different makes were used on the same varieties of cotton. The percentage of lint also varies with the season and the soil on which the crop was grown. In some cases the length of the staple has been accurately measured from samples taken from the seed by hand, while in others the measurements given have been those of the staple after it had been passed through the gin. In several cases where samples of the same variety have been accurately measured the length has been found to vary greatly, owing to differences in soil and season. The size, weight, and color of the seed appear to be of less value in the identification, of varieties than has generally been sup- posed, as either or all of these may vary with differences of soil and with the individual plants from which they have been taken. When inconsistencies have been found in the published descriptions of any variety, it has usually been possible to account for them, and to make due allowances. Eecords which have been extreme and evidently faulty have been ignored, so that the yields and measurements given here are the averages of records which are apparently reliable. It is useless to attempt giving exact characters to such variable plants, but fortunately nearly all of the varieties mentioned in this chapter have been frequently described, and nearly all of them have been grown two or more years under the personal supervision of the writer. In pre- paring the following descriptions it has been the aim to give the general habit and product of each variety, avoiding both the highest and the lowest extremes : AMERICAN VARIETIES. Acme (Allen Acme). — From C. B. Allen, Nanachehaw, Miss. Evi- dently a mixture of some long-staple variety and Sea Island. It is not a hybrid, and does not seem to have any special value. Allen (Allen Silk, Allen Long Staple, Talbot). — Originated by J. B. Allen, Port Gibson, Miss. Plant vigorous, pyramidal, long limbed; bolls large, round, opening very widely, and sometimes allowing the seed cotton to drop; maturing late; lint 28 to 30 percent; staple 30 to 35 mm., fine and silky. This, like the Cook, is easily affected by a change of soil or climate, and produces a longer and better staple when grown in the Yazoo Delta than elsewhere. Aired. — History unknown, but evidently of the Eio Grande type. Reported only once, from Mississippi. Alvarado. — History unknown; appears to be Peterkin. Eeported only once, from Georgia. Audrey Peterkin. (See PeterMn.) i Alabama College Sta. Bui. 33. CULTIVATED VARIETIES OF COTTON. 199 Bahama. (See Texas Storm Proof.) Bailey. — Originated by T. J. King, Louisburg, X. 0. Plant of medium size, early and prolific for a long-staple variety; lint 28 to 30 per cent; staple 2S to 32 mm. An excellent long-staple variety for uplands. Banana (Cluster, Hogan, Prout). — Newspaper writers of 1848 to 1850 mention these as being identical, but give no description further than that the bolls were in clusters and that the seed cotton yielded about 31 per cent of lint. The variety seems to have been discarded many years ago. Bancroft Herlong. (See Herlong.) Bancroft Prolific Long Staple. — Origin unknown; advertised by a Georgia seedsman in 1892, and planted at the Louisiana Station, where its yield was much below the average. Bancroft Prolific Herlong. (See Herlong.) Barnes. — One of the older varieties. Plant vigorous, short limbed, and inclining to cluster, similar to Herlong in habit ; bolls above medium size, maturing late. Probably not now in cultivation. Barnett. — From Alabama; origin unknown. Plant tall and slender; limbs short; bolls medium size, rounded, not maturing early; lint 30 to 32 per cent, staple 23 to 25 mm. Beported only from Alabama. Bates Big Boll. — Originated by B. Bates, Jackson Station, S. C, who developed it by repeated selections of choice plants belonging to the Bio Grande type. Plant vigorous, very symmetrical, well branched; bolls rather large, not maturing early; lint 33 to 35 per cent, staple 21 to 27 mm. In 1892 this gave the largest yield of any of the 25 varieties grown at the Georgia Station, and in 1893 ranked fifth among 26 varie- ties grown at the Mississippi Station. Bates Favorite. — Of the same origin as Bates Big Boll. Plant very vigorous, branching widely; bolls medium size, maturing late; lint 30 to 32 per cent, staple 24 to 27 mm. Belle Creole. — The immediate ancestor of the Jethro and described by Col. H. W. Vick, of Vicusburg, Miss., as follows: " Stalk large, tall, and productive; boll large and long; seed commonly flat on one side with an indentation; lint abundant, long, firm, silky, soft, lustrous, and beyond measure more oily than any cotton I have seen." Ben Smith (Ben Smith Choice, Bush, Smith Standard).— From B. F. Smith, Bedwine, La. Plant strong, widely pyramidal; bolls medium size, usually 2 at each joint, not maturing early; lint 32 to 33 per cent, staple 23 to 20 mm. This was grown in Alabama and Georgia many years ago under the name of "Bush,"' and is probably descended from the Purple Stalk or Bed Leaf, which was common in those States from 1845 to 1850. Big Boll — From California; history unknown, but supposed to be of Texan origin. Plant medium size; limbs rather long; bolls large, oblong, maturing late; lint 34 to 35 per cent, staple 25 to 28 mm. Black Seed. — A name applied both to Sea Island varieties and to upland varieties having a smooth seed. 200 THE COTTON PLANT. Bob, Bob Silk, Bob White. (See Osier.) Bolivar County. — A Louisiana variety of the Storm Proof type, maturing early, with 29 to 30 per cent of lint. Borden Prolific. — Mentioned in South Carolina reports, but no descriptions are given. Boyd Prolific. — One of the oldest of the improved varieties, having been common in Mississippi in 1847, and is the parent stock of many cluster varieties of recent introduction. The originator, Mr. Boyd, said that it was grown from a single plant found in a field of "common" cotton. Plant upright, slender, moderately vigorous, short limbed; bolls small, round, in clusters, medium in time of ripening; lint 30 to 32 per cent, staple 20 to 24 mm. Brady. — Mentioned in the report of the North Carolina Station for 1887, but no description given. Bwgg Long Staple. — Prom T. J. King, Louisburg, 1ST. C. This has every appearance of being a true hybrid between Gossypium herbaceum and Gossypiuni barbadense, but as it was grown from a single stalk found in the field of an ignorant negro its parentage is unknown. Plant very vigorous, well branched; bolls large, oblong, maturing late; lint about 30 per cent, and the staple extremely variable in length, the bulk of the fibers being only about 35 mm., while a few — perhaps 5 per cent of those in each boll — reach a length of fully 75 mm. Owing to its mixed character the staple is classed commercially as " short." Brannon (Little Brannon). — One of the older varieties, originating in Texas many years ago. Plant medium in growth, well branched; limbs short jointed; bolls small, medium in time of ripening; lint 32 to 35 per cent, staple 18 to 22 mm. Belongs to the Rio Grande type. Brazier Peterkin. (See Peterkin.) Brooks Improved. — A local Louisiana variety, maturing early, with 31 to 33 per cent of lint and a short staple. Brown. — Originated in Copiah County, Miss., prior to 1848, and said to have been developed by selections from the Sugar Loaf, an early maturing, short- staple variety. Bush. (See Ben Smith.) Carolina Pride. (See Early Carolina.) Catacaos. (See Peruvian.) Catawba. — From W. R. Davis, Landsford, S. C. A local variety, maturing late, with 35 to 3G per cent of lint and staple 23 to 25 mm. Apparently from the Peterkin. Chambers. — A South Carolina variety of unknown origin, yielding 31 to 32 per cent of lint, with a staple 22 to 25 mm. Belongs to the Her- long type. Champion Cluster. — Plant very vigorous, with long limbs; bolls large, oblong, maturing late; lint 30 to 31 percent; staple 25 to 28 mm. This name is misleading, as the variety does not belong to the cluster type, but is much more like the Mammoth Prolific. CULTIVATED VARIETIES OF COTTON. 201 Cherry Cluster (Cherry). — Originated in South Carolina. Plant of medium growth, cone shaped; limbs of medium length; bolls small, round, clustered, maturing early; lint 30 to 32 per cent; staple 18 to 22 mm. Belongs to the same type as the Dickson. Cherry Long Staple Prolific. — Nearly, if not quite, identical with Cherry Cluster. Originated in the same locality. Cluster. (See Banana.) Cobweb (Spider Web). — Originated in 18S1 by W. E. Collins, Mayers- ville, Miss., and claimed to be hybrid between Peeler and an Egyptian variety of Gossypium barbadensc. Plant very vigorous, long limbed; bolls large, somewhat pointed, maturing late; lint 28 to 29 per cent; staple 35 to 40 mm., very fine and silky. Cochran (Cochran Extra Prolific, Cochran Short Limbed Prolific). — Originated in Georgia by selections from Peerless stock. Plant a moderate grower, slender, short limbed; bolls medium size, round; lint 32 to 33 per cent; staple 35 to 40 mm. Differs but little from the Peerless. Colthorp Eureka. (See Eureka.) Colthorp Prickle. — The name apparently a mistake for Colthorp Pride. Colthorp Pride. — From A. S. Colthorp, Milliken Bend, La. Originally grown from a few seeds picked up from the floor of a country store. Plant vigorous, upright, pyramidal; bolls large, oval, maturing late; lint 28 to 30 per cent; staple 28 to 32 mm., fine and silky; seed small and many of them black. Prolific for a long-staple variety. Cook. — Originated by W. A. Cook, of Newman, Miss., from a single stalk found in a field of " common" cotton in 1884. Plant very vigor- ous and prolific; limbs irregular, not long; bolls large and long, some- times 24 inches in length, maturing late; lint 26 to 28 per cent; staple 35 to 40 mm. Similar to the Allen, and one of the best varieties for rich, low ground. Cox Royal Arch Silk. — A local variety from Georgia, similar to the Ozier Silk. Crawford Peerless (Crawford Premium). — From J. M. Crawford, Colum- bia, S. C. Developed by selections from the Peerless, and practically identical with that variety except that the bolls are usually clustered. Crossland. (See Peterkin.) Dalkeith Eureka. (See Eureka.) Dean. — A local South Carolina variety. Bearing (Dearing Prolific, Dearing Small Seed). — Very similar to Herlong. The Southern Live Stock Journal (1887)1 gives an inter view with Dr. J. J. Dearing, Columbus, Ga., presumably the originator or disseminator of this variety. The writer claims to have made all his selections through a number of years with special reference to the per cent of lint, and claims to have attained finally 45.28 per cent of lint "under the test of Prof. H. C. White, State chemist of Georgia." lU. S. Dept. Agr. Library ScrapLook, Cotton and Sugar, pp. 2, 3. 202 THE COTTON PLANT. Diamond. — Another of the varieties originated by Col. II. W. Vick; of the Bio Grande type, and differing very slightly from that form. Dickson (Dixon, Dickson Cluster, Dickson Improved, Simpson). — Mr. Capers Dickson, of Oxford, (la., writes that this " was originated in this place in 1857 or 1858 by my father, Mr. David Dickson, and was developed from Boyd Prolific cotton by several successive years selec- tions from the seed of that variety. ~No crossing was practiced, and the variety was entirely the result of careful selections each year. The variety has ever since been kept up in the same way." Plants vigor- ous, well branched, pyramidal; limbs short; bolls medium to large, round, clustered, maturing rather early; lint 31 to 32 per cent; staple 23 to 20 mm. One of the most popular cluster varieties. Drake Cluster. — Originated in 1882 by It. W. Drake, of Greensboro, Ala., and was developed from the Peerless "by selecting the earli- est.and best matured bolls from the earliest and most prolific plants." It resembles Peerless in every way excepting that it matures some- what earlier and the bolls are more clustered. Lint 31 to 33 per cent, staple 22 to 25 mm. One of the most popular varieties for Uplands. Drought Proof. (See Texas Storm Proof.) Duncan (Duncan Mammoth). — From F. M. Duncan, Dallas, Ga. A late-ripening, large-boll, long-staple variety, similar to Mammoth Prolific. Early Carolina (Extra Early Carolina, Carolina Pride, South Caro- lina Pride). — An early ripening variety which has been developed by selections from the Dickson and which is very similar to that form. East (East Improved Georgia). — A long-staple variety similar to the Allen, but maturing a little earlier and having a little shorter staple. Lint 31 to 32 per cent. Ellsworth.— Prom W. H". Ellsworth, Wallace, K C. Plant usually with long and spreading limbs; bolls large, oblono-, maturing late; lint 30 to 33 per cent, staple 21 to 21 mm. This is exceedingly variable in its habit of growth and in the character of its lint. Apparently a mixture of two or three varieties. Ethridge. — From W. B. Ethridge, Downsville, La. Grown from a single plant found in a field of some other variety. Plant of fair size; limbs long and spreading; maturing late; staple fine and silky, 28 to 30 mm. ; seed black. Eureka (Colthorp Eureka, Dalkeith Eureka, Humphrey Eureka). — Originated from a single stalk found on a Louisiana plantation many years ago, and for some time was called Mand Atkins. The plant is very vigorous and prolific; limbs of medium length; bolls rather large, oblong, not maturing earl}', holding the seed well in wet weather; lint 28 to 30 per cent, staple 35 to 40 mm., very fine, strong, and silky; seed quite small and sometimes black. One of the most popular long-staple varieties. CULTIVATED VAEIETIES OF COTTON. 203 Excelsior. — Originated in 1883 by C. R. Ezell, Eatonton, Ga.,by selec- tions from the New Era. Similar to the Peterkin, though "with bolls a trifle larger. Lint 33 to 35 per cent, staple 26 to 30 mm. Farrar Forked Leaf. (See OTcra.) Ferrell Prolific. — From C. B. Ferrell, Montgomery, Ala. Plant me- dium size, with very long and straggling limbs ; very prolific ; bolls large, oblong; lint 28 to 30 per cent, staple 30 to 35 mm. Similar to Mammoth Prolific. Georgia Prolific (Georgia Upland). — Names which are applied to a number of upland short-staple varieties of the Peterkin and Herlong types. Garuer. — A local Alabama variety of the Eio Grande type- Gold Dust. (See King.) Grayson Early Prolific. — From W. B. Grayson, Grayson, La. Plant medium in size; limbs short, not spreading widely; very prolific; bolls medium in size, somewhat clustered, ripening early; lint 34 to 36 per cent, staple 22 to 25 mm. Resembles Peterkin, but matures earlier. Griffin. — This remarkable variety was originated by John Griffin, of Greenville, Miss., who developed it by repeated and persistent selec- tions from some unknown long-staple variety. Plant vigorous, with a pale-green leaf, prolific; bolls large, medium in time of maturing; lint 28 to 29 per cent, staple very fine aud silky, occasional fibers 70 to 75 mm. The longest and finest staple we have found. Gunn. — From C. L. Gunn, Temple, Miss. A local variety which does not differ materially from the Rio Grande type. Hawkins (Hawkins Extra Prolific). — Originated by B. W. Hawkins, Nona, Ga., from a single stalk in a field where mixed seed of Boyd Prolific, Herlong, and New Era had been planted. From the prod- uct of this plant repeated selections were made, until the variety had assumed a fairly constant form. The plants are very vigorous, well branched, pyramidal, prolific; bolls medium in size, roundish, medium or early iu time of maturing; lint 32 to 34 per cent, staple 18 to 22 mm. Mays China. — From J. W. Hays, Hays Landing, Miss. Very similar to the Allen, and perhaps the same. Herlong (Bancroft Herlong, Jones Herlong, etc.). — From E. Bancroft, Athens, Ga. Plant medium in size, well branched, pyramidal, very prolific; bolls medium in size, round, maturing rather late; lint 30 to 32 per cent, staple 20 to 25 mm. A semicluster variety, very popular in Georgia and Alabama. This name is sometimes incorrectly applied to a long-staple variety. Hightower. — A local Alabama variety, strong growing, bolls very large, and staple of medium length. Hilliard. — From W. A. Hilliard, Bowersville, Ga. Of the Rio Grande type, and not differing essentially from Peterkin. Hogan. (See Banana.) 204 THE COTTON PLANT. Hollingshead. — One of the oldest varieties of which we have any record, it having been grown in the Oarolinas in 1818, It was sup- posed to be of Mexican origin, and seems now to have disappeared. Howell. — A local Louisiana variety, very similar to Peterkin, but perhaps a little earlier in maturing. Humphrey Eureka. (See Eurelca.) Hunnicutt (Hunnicutt Choice). — Originated by Prof. J. B. Hunnicutt, Athens, Ga., who developed it by mixing seed of Bates, Boyd Pro- lific, Herlong, Truitt, and other varieties, and planting them in the same field. From the product oi" these seeds stalks approaching nearest the originator's ideal form were selected for further similar selections which were continued through several years. The plant is large and well branched, branches spreading, prolific; bolls of medium size, roundish, maturing early; lint 30 to 32 per cent, staple 22 to 25 mm. Improved Long Staple. (See Jones Long Staple.) Improved Prolific. — From A. Borden, Goldsboro, ET. C. A local variety, differing but little from the Herlong. J. 0. Cool\ — Evidently a descendant of the old Purple Stalk type, and apparently identical with the Ben Smith. Jenkins (Jenkins Poor Man's Friend). — Originated by J. F. Jenkins, Natchez, Miss., by repeated selections from Brannon. Plant strong, pyramidal, prolific; bolls medium in size, oval, maturing early; lint 34 to 36 per cent, staple 22 to 25 mm. One of the best of the Rio Grande type. Jethro (McBride Silk).— Originated by Col. H. W. Vick, of Vicksburg, Miss., between 1830 and 1840, by selections from the Belle Creole, and sent to J. V. Jones, of Herndon, Ga., in 184G, with a request that it be called by its present name. We have been unable to find any descrip- tion of this variety and can not learn that it is now in cultivation. It is the parent stock of Jones Long Staple, Six Oaks, and a number of other similar varieties. Jones Herlong. (See Herlong.) Jones Improved (Jones Improved Prolific). — Plant of medium size, limbs short and spreading, not very prolific; bolls large, roundish, maturing late; lint 30 to 32 per cent, staple 20 to 24 mm. Jones Long Staple (Improved Long Staple, Bichardson Improved). — Originated by J. H. Jones, Herndon, Ga. Plant large; limbs long and spreading, prolific; boils large, oval, pointed, maturing medium or late; lint 29 to 30 per cent, staple 30 to 34 mm. One of the many descendants of the Jethro, and one of the most popular long- staple varieties for the middle and southern parts of the cotton-growing region. Jones No, 1. — A local Alabama variety of the Eio Grande type, pro- ducing 33 to 34 per cent of lint, with a staple 18 to 22 mm. lowers (Jowers Improved). — Very similar to Peterkin, and probably the same. CULTIVATED VARIETIES OF COTTON. 205 Jumbo. — Originated by B. W. Hawkins, ISTona, Ga., by selections from the Hawkins, and similar to that variety, though more prolific. Kelly. (See Marston.) Kieth. — From Alabama. Plant tall, pyramidal; limbs short jointed, prolific; bolls medium in size, roundish, not clustered, maturing early; lint 30 to 32 per cent, staple 24 to 27 mm. King (Gold Dust, King Improved, Tennessee Gold Dust). — Originated by T. J. King, Louisburg, X. C. Plant of medium size, pyramidal, well branched, very prolific; bolls small, roundish, all maturing early; lint 32 to 34 per cent, staple 25 to 28 m«i. ; seeds small. The fact that this variety matures its entire crop very early makes it one of the most desirable sorts for the northern cotton belt. Lewis Prolific. (See Sugar Loaf.) Little Brannon. (See Brannon.) Louisiana. — A name which, like the Georgia Prolific, is applied to a large number of upland short-staple varieties. Magruder Marvel. — Originated by I. D. Magruder, Eussum, Miss., and is a cross secured by fertilizing the flowers of Barnes Cluster with pollen from Golden Prolific. Plant pyramidal; limbs abundant and short jointed; bolls small, round, somewhat clustered, maturing early for a cluster variety; lint 31 to 33 per cent, staple 25 to 30 mm. Magruder XL. — Originated by I. D. Magruder and secured by ferti- lizing Peterkin with pollen from the Allen, and in the after fixation of the type by keeping the selection of plants for seed as near as possi- ble to the Peterkin in production and habit, and at the same time preserving the longer staple secured from the Allen. Early and pro- lific; lint 32 to 34 per cent, staple 25 to 30 mm. Mallius Prolific. — A local Louisiana variety. Mammoth Cluster. — From Georgia. One of the older varieties and of unknown origin. Quite similar to Champion Cluster. Mammoth Prolific. — Another Georgia variety of unknown origin. Plant very strong, well branched, not very prolific; bolls very large, oblong, maturing very late; lint 30 to 32 per cent, staple 26 to 30 mm. Yery similar to Duncan, and perhaps the same. Marston (Kelly). — From Louisiana. Plant of medium growth; limbs short, prolific; bolls fair size, round, maturing late; lint 30 to 31 per cent, staple 26 to 30 mm. Martin Prolific. — A Louisiana variety, and apparently the same as Marston. Mastodon. — This name occurs in a list of the varieties grown in Mis- sissippi in 1850, but no description is given, and the variety seems to have disappeared. Matthews. — From J. A. Matthews, Holly Springs, Miss., who saved the seed from a stray plant found in his garden. Plant very vigorous, pyramidal, with limbs from near the ground; limbs short jointed, very prolific; bolls large, ovate, pointed, maturing early; lint 29 to 30 per 206 THE COTTON PLANT. cent, staple 35 to 40 mm. This has given remarkably good yields for a long-staple variety in the northern cotton belt. Mattis. — From 0. F. Mattis, Learned, Miss. Developed by repeated selections from some unknown variety. Plant vigorous; limbs long, short jointed, prolific; bolls clustered, medium in size, maturing rather late; lint 30 to 32 per cent, staple 25 to 30 mm. Maxey (S. B. Maxey, Meyers, Meyers Texas). — Of unknown origin, but introduced in Texas by Hon. S. B. Maxey. Plants medium in size, well branched, prolific; bolls large, roundish; lint 31 to 32 per cent, staple 30 to 35 mm. Esteemed more highly on bottom lands than on uplands. McAllister Peerless. (See Peerless.) McBride. (See Jethro.) McCall. — A variety which was somewhat popular in South Carolina about 18S0, but which seems now to have been dropped. It belonged to the same group as Truitt and other similar varieties. Mclver, — A local South Carolina variety similar to McCall. Mexican. — One of the oldest known varieties, having been brought from the City of Mexico to Natchez, Miss., by Walter Burling in 1800, and introduced in South Carolina as early as 1816. It was from this stock that by far the larger proportion of our short and medium staple varieties have been developed. Mexican Burr. — An old name for the varieties of "Mexican" which produced bolls in clusters, and the original source of many of the pres- ent cluster varieties. Meyers. (See Maxey.) Minter (Minter Prolific). — From J. E. Minter, Laurens, S. C. Origi- nated about 1870 by selections from some unknown variety. Plants large, branching low and widely, prolific; bolls medium in size, round or oval, maturing late; lint 30 to 32 per cent, staple 22 to 25 mm. Quite similar to Herlong. Moina. — Originated by J. M. Taylor, of North Carolina, about 1873, and described as follows: "This variety is remarkable for the number of its limbs, but they grow too long and slender. Its fiber, in length and fineness, is said to surpass the Peeler. It bears bolls abun- dantly, but is troublesome to pick and gin, so that its culture is gener- ally abandoned in this section." Money Bush. — Mentioned as one of the varieties growing in Missis- sippi in 1850, and probably the same as Banana. Moon. — From J. Moon, Peytonville, Ark. Originated about 1875 from a single plant showing unusually good staple. Plant strong; limbs long and spreading; bolls large, oval, medium in time of matur- ing; lint 31 to 33 per cent, staple very strong and silky, 30 to 35 mm. Multibolus. — This was grown in Mississippi in 1849, and some- times called "Sugar Loaf." It was a cluster variety which has now disappeared. CULTIVATED VARIETIES OF COTTON. 207 Multifiora. — Grown in Alabama in 1849, and said to be similar to tlie Banana, but with larger clusters of bolls and lighter-colored seeds. Oats. — Grown and named at the Louisiana Station from seed of an unknown variety purchased at a public gin in 1889. Plant vigorous, sugar loaf in shape, very prolific, maturing early ; lint 32 to 31 per cent; staple 20 to 25 mm. Perhaps the same as King. Okra (Okra Leaf, Farrar Forked Leaf). — One of the older varieties, and mentioned by Southern writers as early as 1837, when it was quite common and somewhat popular, but it soon disappeared from general cultivation. About 1870, C. A. Alexander, of Washington, Ga., found a single stalk in his field, and from its product the variety was again disseminated quite widely from 1885 to 1890, but its culture seems less general now than five years ago. The plant is of medium growth; limbs short and upright; leaves with very narrow lobes; bolls clus- tered, small, round, maturing early; lint 32 to 31 per cent; staple 21 to 20 mm. The small amount of foliage produced by this variety is said to preserve it from attacks by worms and to hasten the complete maturity of the crop in wet seasons. Ozier (Ozier Silk, Bob, Bob Silk, Bob White, Tennessee Silk). — From J. D. Ozier, Corinth, Miss. Plants medium size, pyramidal; limbs rather short, moderately prolific ; bolls medium in size, oval, ripening early; lint 30 to 32 per cent; staple 25 to 28 mm. Quite popular from 1880 to 1890, but less so now. Pearce. — Introduced by T. J. King in 1887, and described as an early maturing sort, producing 32 to 33 per cent of lint, but does not appear to have been widely disseminated. Peeler. — Originated in Warren County, Miss., about 1864. Plant very large and vigorous, branching widely; bolls large, maturing late; lint 30 to 32 per cent; staple very strong and silky, 25 to 28 mm. One of the most widely cultivated varieties. Peerless (Crawford Premium, Crawford Peerless, McAllister Peerless, Sutton Peerless, The Premium). — Probably of Georgia origin. Plant medium, well branched, pyramidal; bolls small or medium in size, round, sometimes clustered, maturing early; lint 32 to 33 per cent; staple 23 to 27 mm. One of the best of the upland varieties. Peruvian (Catacaos). — A South American variety of Gossypium arbo- reum, which never matures fruit in the LTnited States. Peterkin (Audrey Peterkin, Brazier Peterkin, Crosslaud, Texas Wood, Wise).— Originated by J. A. Peterkin, Fort Motte, S. C, about 1870. Originally a variety with smooth, black seeds, and producing nearly 50 per cent of lint, and developed to its present form by repeated selec- tions of seed from the most prolific plants. Plants of medium size, well branched; limbst short jointed; bolls medium in size, oval, not clustered, not maturing very early; lint 34 to 36 per cent; staple 22 to 25 mm. Seed occasionally black and smooth. There are very few varieties which yield so large a percentage of lint, and this is one of the best of the Eio Grande type. 208 THE COTTON PLANT. Peterlcin Limb Cluster (Peterkin New Cluster). — Developed by J. A. Peterkin through selections from the Peterkin, and similar to that variety excepting that the bolls are somewhat clustered. Petit Gulf. — One of the oldest varieties ; originated by Col. H. W.Vick, the originator of the Jethro, about 1840. In 1848-40 large quanti- ties of the seed were sent to Georgia and Alabama, and as the ship- ments were all made from Petit Gulf the variety became known under that name. The plant is large, long limbed, and long jointed, not very prolific; bolls medium in size, ovate, not maturing early; lint 30 to 32 per cent; staple 22 to 25 mm. Boyd Prolific and Dickson are i>robably descended from this variety. Pittman (Pittman Extra Prolific, Pittman Improved). — An early maturing, short-limbed, cluster variety, from Louisiana, and which does not differ essentially from Dickson. Pitt Prolific. — Mentioned as having been grown in Mississippi in 1848, and doubtfully distinct from Banana. ISTo longer in cultivation under that name. Pollock. — Originated in 1800 by W. A. Pollock, Greenville, Miss., by fertilizing some unknown long-staple variety with pollen from Peer- less. A cluster variety maturing a little later than the Peerless, with a staple 35 to 40 mm. Poor Maris Belief. — A California variety, not distinguishable from Peterkin. Prolific. (See Sugar Loaf.) Prout. (See Banana.) Queen (Southern Queen). — A local Arkansas variety, very similar to Peterkin, and probably the same. Barneses. — An old variety, apparently a form of Peerless. Eichardson Improved. (See Jones Long Staple.) Bio Grande. — An early name for the original form of many of the upland short-staple varieties, yielding 34 to 36 per cent of lint, with a staple 18 to 22 mm. Bod Smith 25 Gent. — Grown in Mississippi in 1849, and described as having "limbs long and slender, bolls slim, very prolific." Not now known. Boe Early. — A local Louisiana variety, maturing early, with 28 to 30 per cent of lint, and a staple 25 to 30 mm. S. B. Maxey. (See Maxey.) Sea Lsland. — This is a native of the West Indies and Central America, and was one of the first varieties cultivated in the United States. In the limited region where it can be grown it is the most desirable variety, as it produces a very long and fine staple, which commands the highest price, but it is seldom profitable when grown more than 50 miles from the Atlantic coast. Shine Parly. — From J. A. Shine, Shine, N. C. An early maturing variety of the Rio Grande type, and differing but little from Peterkin. Silk. (See Jethro.) CULTIVATED VARIETIES OF COTTON. 209 Simpson. (See Dickson.) Six Oaks. — Originated by J. V. Jones, Herndon, Ga,, the original form being the Jethro, which was sent from Mississippi in 1840. It is similar to the Jones Long Staple, except that the plant is less vig- orous, the bolls are not quite so large, and the seed are smooth and black. Lint 28 to 30 per cent, staple 35 to 40 mm. Smith Standard. (See Ben Smith.) South Carolina Pride. (See Early Carolina.) Southern Hope. — Originated many years ago by Col. F. Eobieu, of Louisiana, from seed said to have come from Peru. Plant pyramidal; limbs strong and straight, prolific; bolls large, pointed, maturing rather late; lint 30 to 32 per cent, staple 28 to 32 mm. One of the best types for the southern cotton belt, but maturing too late for northern latitudes. Spider Web. (See Cobweb.) Storm Proof. (See Texas Storm Proof.) Sugar Loaf (Lewis Prolific, Prolific, Vick 100 Seed). — This is said to have originated in Yalobusha County, Miss., in 1843-44, and to have come from Mexican seed. A cluster variety not now in cultivation. Sutton Peerless. (See Peerless.) Talbot, (See Allen.) Tarver. — Grown in Alabama as early as 1848, and probably the same as Sugar Loaf. Taylor. — This name is given to at least two distinct varieties, the one known in Alabama under that name having a small boll and a short staple, while that from South Carolina has a large boll and a long staple. In both States the name seems to be only local. The original Taylor of Alabama was claimed to be a cross between the Purple Leaf and Boyd Prolific, showing principally the characters of the former. Tennessee Gold Dust. (See King.) Tennessee Silk. (See Ozier.) Texas Storm Proof (Bahama, Drought Proof, Storm Proof). — Prom W. J. Sinilie, Baileyville, Tex. Plant tall, with slender and often drooping limbs, not very prolific; bolls large, pointed, maturing late; lint 33 to 35 per cent, staple 23 to 2G mm. The matured seed cotton does not fall from the bolls as readily as in most varieties, and hence its name of " Storm Proof." Texas Wood. — Prom D. F. Miles, Marion, S. C. Not distinguishable from Peterkiu, and probably identical with that variety. The Premium. (See Peerless.) Truitt Premium (Truitt Improved). — Originated by G. W. Truitt, Lagrange, Ga, Plants large; limbs long and spreading, prolific; bolls very large, roundish, maturing late; lint 30 to 32 per cent, staple 22 to 25 mm. Very similar to Duncan Mammoth and Mammoth Prolific, Vick 100 Seed. (See Sugar Loaf.) 1993— No. 33 14 210 THE COTTON PLANT. Welbom Pet. — Originated by Jeff Welborn, New Boston, Tex. Developed from selected plants in a field of Barnes, Jones Big Boll (Jones Long Staple?), and Zellner. Plant erect, slender; limbs short and numerous, very prolific; bolls round, medium in size, clustered, maturing early; lint 31 to 32 per cent, staple 22 to 25 mm. This variety lias but little foliage in proportion to the size of the plant, which many cultivators claim is an advantage in hastening early and uniform maturity. One of the best known cluster varieties. Williams. — An old short-staple variety yielding 33 to 35 per cent of lint and probably identical with Peterkin. Williamson. — From E. M. Williamson, Dovesville, S. C. Plant not large; limbs short, prolific; bolls small, round, maturing early; lint 30 to 31 per cent, staple 22 to 25 mm. Willimantic. — Very similar to Duncan Mammoth and Mammoth Pro- lific. Willis. — From J. B. Allen,' Port Gibson, Miss. Much like Allen in growth, with lint 20 to 30 per cent, and a staple 33 to 37 mm. Wimberly. — From F. D. Wimberly, Billiards, Ga. Can not be dis- tinguished from Duncan. Wise. (See Peterkin.) Wonderful (Jones Wonderful).-— From J. H. Jones, the originator of several other varieties bearing his name. Tliis is similar to the Jones Long Staple, but has a larger boll and a smaller seed, with a longer and finer staple. Bulletin 20 of the Georgia Station says "it is an excel- lent type of the upland long-staple varieties, and is more productive and of better staple than any other of the class tested on the station." Lint 28 to 30 per cent, staple 35 to 40 mm. Zellner. — Probably developed by selections from Boyd Prolific. Plant small to medium, limbs short, prolific; bolls medium or small, round, maturing early; lint 30 to 31 per cent, staple 20 to 25 mm. Very much like Dickson. In the preparation of the foregoing descriptions great care has been exercised to avoid duplication of varieties under different names, though that has probably been done in a few instances, and no names have been given as synonyms excepting when it has been impossible to separate the varieties either by description or by personal observation. As showing the instability of varieties it is interesting to note that in the report on cotton of the Tenth Census, 58 varietal names are mentioned. Of these, we have not been able to find that more than 15 have been used during the last ten years, and only G of the varieties popular in 1880 are still in common cultivation. Those which have stood the test of the fourteen years since 1880 are Boyd Prolific, Dick- son, Herlong, Peeler, Petit Gulf, and Texas Storm Proof. FOREIGN VARIETIES. Several varieties of cotton from Japan, Russia, Egypt, and India have been planted in the United States during the last ten years, but CULTIVATED VARIETIES OF COTTON. 211 none of them has proved suited to the conditions in this country. The varieties received from Japan have produced very dwarf plants, with small bolls, very small seeds, and a staple not more than 12 to 15 mm., which has been harsh and woolly. The Russian cottons have been uni- formly light in yield, short and weak in staple, and usually somewhat colored. The Egyptian varieties are closely related to the Sea Island, and produce an immense growth of stalk.1 Afifi and Bamia are the two varieties which have been most widely tested, but neither has proved to be profitable. Both produce a very long and fine staple, but mature too late for our climate. Seeds of both these varieties were distributed quite freely by the Department in 1892, and since then a number of hybrids between them and some of the upland American varieties have been reported winch promise to have considerable value, especially in the southern part of the cotton region. The Indian vari- eties which have been received have been of two distinct types. One is much like the Japanese varieties in leaf, boll, and lint, but produces a large and spreading plant which bears a very light crop. The other type is evidently descended from the American seed which was sent to India in 1844, and which has become quite common in that country. It is interesting to note that this American cotton which has been grown in India for fifty years has come back to this country practically unchanged, and can not now be distinguished from the Petit Gulf, which was so common in this country from 1830 to 1850. ORIGINATION OF VARIETIES. Of course the origin of many of the varieties has been lost in obscu- rity, but from what has been gathered it appears that the most frequent methods by which new varieties have been originated have been by (1) the selection of individual plants for the original stock; (2) the saving of seed from the earliest maturing bolls, and planting them (usually) on soil which had been highly fertilized ; (3) cross fertilization, and (4) the very simple process of changing the name. PLANT SELECTION. Every observant planter has noticed the great differences which may be seen among the plants in any field, even though the seed used may all have come from the same source. The different plants will vary in height and vigor of stalks, in length and direction of limbs, in size and sliape of leaves, and in the arrangement of bolls on the stalk. Some plants will mature all their bolls within a few days of each other, while other plants will mature slowly and yield successive pickings through several weeks. Any one of these characteristics may be, to a certain extent, strengthened and perpetuated by the saving of seed from plants having the desired characters most strongly developed, and afterwards 1 See also article on history of cotton, p. 47. 212 THE COTTON PLANT. growing tlieni in a field so far removed from other cotton that there will be no clanger from cross fertilization. If this process is repeated for a few years, selecting- seed each year from those stalks which are nearest the ideal form, the variety will become more and more fixed with each succeeding crop, and each year the entire crop will approacli nearer to the desired standard. It has been by this process that many of the more strongly marked varieties like Cook, Dickson, and Welborn Pet have been produced. A bulletin of the Louisiana Station ] says : First, determine what you want — have a fixed standard in view. Then goto your plants, whatever they may he, and select carefully that one which comes nearest to filling these requirements. Then carefully gather the seeds of that plant, and in the proper season plant them. Again exercise the same care in selecting your seed and repeat the operation until you finally reach the ohject required. It takes time for this, and there are frequent disappointments. Do not misunderstand us as recommending every farmer to go into tho business of originating varieties. At the most, whether it be by selection or by hybridizing, it requires more time and patience than many of us possess. But wo do urge upon each and all the importance of exercising the greatest care iu the selection of their seed. The day is not far distant, in fact is even here, when the kind of seed used is quite an item in the profit and loss account of each farmer, and he who neglects it is bound, sooner or later, to go to the wall to make way for a more careful successor. SAVING SEED FROM EARLY MATURING ROLLS. The second method named, that of saving seed from the earliest matur- ing bolls, the "first pickings," is practiced largely, and it is to this process that we owe many of the widely known varieties like Drake Cluster, Jethro, and Southern Hope. The higher fertilization and cultivation which seed saved in this manner usually receive also tends to lengthen the fiber, make the plants more productive, and so assists still further variation from the original type. Varieties produced in this manner, though usually superior to the original stock, are not strongly marked, and the improvement is less permanent than is that which may be pro- duced by the selection of individual plants. In fact, these should hardly be recognized as distinct varieties, and in the naming of such the word "improved" is often prefixed to the name of the parent variety. CROSS FERTILIZATION. When it is desired to combine the good qualities of two varieties in a single stock, the work can be accomplished best by cross fertilization, and it is more than probable that a large proportion of the varieties which have come from single plants have really originated accidentally in this manner. It is equally true that many of the so-called "crosses" and "hybrids" are not such, but are merely sports. Although cross fertilization is the surest method for the production of new varieties, it is largely work in the dark, as the plants resulting from the crosses may fail to show the good qualities of either parent and have all the weak points of both. Out of a hundred crosses which 1 Louisiana Sta. Bui. 2fr CULTIVATED VARIETIES OF COTTON. 213 may be grown, it is .seldom that more than one or two plants will show the combination desired, and even when a promising plant does appear its character is not yet fixed, and several generations must be grown before it will assume its permanent form and demonstrate its real value. Although the plants from a single line of crosses, as fertilizing Peterkin with Allen, will vary widely, still it is a general rule that the character and habit of the future plant will be more like those of the female parent, while the fruit, the boll and its contents, will be more like those of the male parent. The flower of the cottou plant is so large and develops so rapidly that cross fertilization is very easily secured. Flowers which are to be fer- tilized should be among those which are developed early in the season, and should always be those on healthy and vigorous plants. The flow- ers to be operated upon should be selected late in the afternoon, as it is only at that time that they are in the proper condition for the work. One side of the unopened bud should be split lengthwise with a sharp knife having a slender blade, and the stamens removed. The anthers, the fertilizing parts of the stamens, will be found well developed and standing well away from the pistil, though not yet so matured as to be discharging pollen. These can be readily separated from their supports by a few careful raking strokes of the knife, and the emasculated flower should then be inclosed in a paper bag to prevent access of pollen from unknown sources. The following morning the pistil will be fully devel- oped and ready to receive pollen. A freshly opened flower from a healthy plant of the variety which it is desired to use in making the cross is picked and carried to the plant which was treated the previous evening, the bag is removed from the prepared flower, and, by means of a camel's hair brush, pollen is dusted over the end and upper rjart of the pistil. The paper bag is then replaced and allowed to remain two days, after which it should be removed. If more than one variety is used to furnish pollen for different flowers, a tag bearing the name ot the variety from which the pollen was taken should be attached to the limb just below each fertilized- flower. When the fertilized bolls are matured, those from each line of crosses should be saved separately and the seed from each planted separately the next season. The plants resulting from such crosses will vary greatly, even when all come from seeds from a single boll. Any which appear to be wanting in vigor should be destroyed as soon as their weakness is shown, and only healthy and vigorous plants allowed to mature. When the plants approach maturity they should again be examined carefully, and all which show a want of fruitfulness, a tendency to disease, or any other undesirable character should be removed. At the same time those plants which approach most nearly to the ideal, i. e., those which are strong and vigorous, which produce the largest number of bolls, and which have the longest and best fiber and the largest proportion of fiber to seed, should be marked, and the seed from each of these saved 214 THE COTTON PLANT. separately. As the earlier maturing bolls always produce the bet- ter crop, the later ripening bolls from the selected plants should be discarded. The following season the seed from each of the selected plants should be planted separately, and then, and not until then, will the planter be able to form an intelligent estimate of the true value of his crosses. It will be seen that such work requires care, judgment, and patience, but it is the only plan by which the special characteristics of different varieties can be combined in one; it is the only way by which the planter can secure such a definite modification of the plant as he may wish, and it establishes varieties having greater permanency of character than can be produced in any other manner. CHANGE OF NAME. By far the larger part of the names of varieties now in cultivation are simply synonyms of other names. Changes of name are commonly made by using the name of the person from whom seed is purchased, giving a new name to an old variety for advertising purposes, substi- tuting a local name for one in general use, or transferring names from one locality to another. Often several varieties receive the same name. IMPROVEMENT AND DETERIORATION OF VARIETIES. * The cultivated cottons of to-day are very different from the original form of Oossyinum herbaeeum, which gave only 28 to 30 per cent of lint, with a staple 20 to 30 mm. long. The proportion of lint has been greatly increased, reaching as high as 36 and even 40 per cent in some varieties, while the length of staple has increased correspondingly, sometimes reaching fully three times its original length. In but few varieties, however, have we secured any marked increase in both j>ercentage of lint and length of staple, and it is in that direction that we should look for improvements in the future. The tendency of the plant to vary from the typical form of any variety has been mentioned on another page, and it should be borne in mind that the natural tendency of this variation will be back toward its origi- nal form rather than in any other direction. Abnormal fruitfulness is always secured at the expense of vitality, and the original form, pro- ducing a smaller amount of fruit, will have the greater vitality, and so a greater prepotency, than will the more fruitful and more desirable, but weaker, plants. If, then, any improved variety is to be kept up to its present standard, constant care in the selection of seed is essential. The seed from all plants which show degeneration should be rejected, and only those from typical plants should be saved. Whenever these precautions are neglected, degeneration is sure to follow, and we soon hear complaints that the variety has "run out." All of our present varieties are artificial, and have been produced by artificial methods, which must be continued as long as the variety is to be maintained. Continued intelligent selection and cross fertilization may work a gradual CULTIVATED VARIETIES OF COTTON. 215 improvement, while neglect in the selection of seed is sure to result in degeneration and reversion to the original form. It is for this reason that planters are so often disappointed with the results secured from new varieties for which extravagant claims have been made by their originators. These new varieties may do all that is claimed for them during the first one or two years, but afterwards, in many cases, dete- riorate rapidly, and their originators are denounced as frauds. The too common method of saving seed for planting is to take a suffi- cient number of bushels just as they come from the gin, or, perhaps, to buy them from an oil mill. Xo attention has been given to the selec- tion of the individual plants from which these seeds came, and those from the poorest, least prolific, and latest maturing are all taken together with those from the best. Seed from the less prolific plants will have greater vitality and so produce stronger plants than those from the more prolific individuals, and when this process is repeated for a few generations it is sure to result in a marked decrease in yield and a deterioration in quality. In speaking of the instability of varie- ties, W. A. Cook says : I can take any of the so-called distinct varieties of cotton and in a few years develop ail the known varieties from it. In other words, they will develop them- selves in the course of time. All that is necessary is to watch the field from year to year, and when a "sport" is noticed, save the seeds and plant them hy themselves. J. Griffin, the originator of the Griffin variety, writes : Since 1868 I have not omitted a single year in my selections, which are guided by length, fineness, prolific tendency, earliness, and, lately, smallness of seed, to humor planters. I can now see that it will take more than the remainder of my life to complete my work, as the variety still improves as at the first. J. H. Jones, of Georgia, says : There is no other plant known in our agriculture which deteriorates so rapidly and requires such a rigid selection of seed to keep it up to a standard as does cotton. The Sixth Eeport of the Mississippi Station says : The particular variety to be selected * * * depends more on the care with which the seed has been selected during the last two' or three years than upon its original source, and without constant care in the selection of seed any variety will soon "run out." The first pickings will give better seed than will the later pickings ; and if the seed be saved from the best stalks only, the practice will soon work a marked improvement in any variety. Even when this can not be done for the entire crop, it will be easy to secure enough of this selected seed to plant a small field, which will produce sufficient seed for the entire crop of the second season. It has been by the following of this plan that nearly all of our best varieties have been developed, and the superiority of any variety is usually a very good measure of the care and judgment which were exercised in selecting the seed plants of the original stock. CLASSIFICATION OF VARIETIES. With so many varieties which differ from each other only slightly, and with their differences those of degree rather than of kind, and all of them exceedingly unstable, it is impossible to classify them by any natural and fixed characters, or even to assign limits to different groups which will not sometimes .separate varieties which resemble each other closely, and any grouping must be purely arbitrary. 216 THE COTTON PLANT. If classified according to the proportion of lint to seed, they may be placed in three groups, as follows : (1) Those having less than 30 per cent of lint, including Allen, Bailey, Bragg, Cobweb, Colthorp Pride, Cook, Eureka, Terrell, Griffin, Jones Long Staple, Matthews, Boe Early, Six Oaks, Wonderful, and Willis. (2) Those having from 30 to 34 per cent of lint. This group will include more than one-half of all our recognized varieties, as follows: Barnett, Bates Favorite, Ben Smith, Boyd Prolific, Chambers, Champion Cluster, Cherry Cluster, Dickson, Drake Cluster, Ellsworth, Hawkins, Herlong, Hunnicutt, Jones Improved, Jones No. I, Kieth, King, Magruder Marvel, Magruder XL, Mammoth Prolific, Marston, Mattis, Maxey, Minter, Moon, Oats, Okra, Ozier, Peeler, Peerless, Petit Gulf, Southern Hope, Truitt Premium, Welborn Pet, Williamson, and Zellner. (3) Those having 34 or more per cent of lint, including Bates Big Boll, Big Boll, Brannon, Catawba, Excelsior, Grayson, Jenkins, Peterkin, Bio Grande, and Texas Storm Proof. If the same varieties are classified according to length of staple only, they may again be placed in three groups, as follows: (4) Those having a staple less than 25 mm., or "short-staple" varie- ties, including Barnett, Ben Smith, Boyd Prolific, Brannon, Catawba, Chambers, Cherry Cluster, Dickson, Drake Cluster, Ellsworth, Grayson, Hawkins, Herlong, Hunnicutt, Jenkins, Jones Improved, Jones' No. 1, Minter, Oats, Peterkin, Petit Gulf, Bio Grande, Texas Storm Proof, Truitt Premium, Welborn Pet, Williamson, and Zellner. (5) Those having a staple of from 25 to 30 mm., or the "medium- staple" varieties, including Bates Big Boll, Bates Favorite, Big Boll, Champion Cluster, Excelsior, Kieth, King, Magruder Marvel, Magruder XL, Mammoth Prolific, Marston, Mattis, Okra, Ozier, Peeler. Peerless, and Boe Early. (6) Those having a staple exceeding 30 mm., or the "long-staple" varieties, including Allen, Bailey, Bragg, Cobweb, Colthorp Pride, Cook, Eureka, Ferrell, Griffin, Jones Long Staple, Matthews, Maxey, Moon, Six Oaks, Southern Hope, Willis, and Wonderful. Group 1 has 15 varieties, of which 14 belong in group 6 and 1 in group 5. The varieties in this group, with the doubtful exception of Boe Early, of which we are able to find but one record, are what are commonly called "long-staple" varieties, and are almost uniformly varieties in which the plant is large and vigorous, the limbs long and spreading, the bolls large, and which usually mature late. Group 2 has 36 varieties, of which 20 belong in group 4, 13 in group 5, and 3 in group G. In this group we have what may be termed the normal type of cotton, with a slight increase in both percentage of lint and length of staple, some varieties showing an improvement in only one direction, but nearly all showing that both the amount and quality of the fiber have been materially bettered by cultivation. Group 3 has 1.0 varieties, of which 7 belong in group 4 and 3 in group CULTIVATED VARIETIES OP COTTON. 217 5. In this group we have varieties in which the percentage of lint has been greatly increased with but little improvement in the character of the staple, nearly every variety in the group having a staple less than 25 mm. in length. Group 4 has 27 varieties, of which 20 belong iu group 2 and 7 in group 3. Group 5" has 17 varieties, of which 13 belong in group 2, 3 in group 3, and 1 in group 1. Group 6 has 17 varieties, of which 13 belong in group 1 and 3 in group 2. It will be observed that the correspondence between group 1, pro- ducing less than 30 per cent of lint, and group 6, having a staple more than 30 millimeters in length, is very close, only 3 of the "long-staple" varieties yielding more than 30 per cent of lint, and only 1 variety which yields less than 30 per cent of lint having a staple less than 30 mm. The correspondence between the other pairs of groups — 2 and 5, 3 and 6 — is less marked, but still shows a strong tendency toward a decrease in length of staple as the percentage of lint is increased. If classified according to time of maturity, the varieties may be grouped as follows : Classification of varieties according to time of maturity. Early. Medium. Late. Bailey. Barnett. Allen. Brooks Improved. Bates Big Boll. Barnes. Cherry Cluster. Ben Smith. Bates Favorite. Dickson. Boyd Prolific. Brag Long Staple. Drake Cluster. Brannon. Catawba. Early Carolina. East. Champion Cluster. Grayson Early Prolific. Eureka. Cobweb. Hiinnicutt. Griffith. Colthorp Pride. Jenkins. Hawkins. Cook. Kieth. Herlong. Ellsworth. King. Jones Long Staple. Magruder Marvel. Ethridge. Matthews. Jones Improved. Oats. Mattis. Mammoth Prolific. Okra. Moon. Marston. Ozier. Peterkin. Minter. Peerless1, Peterkin Cluster. Peeler. Pittinan. Petit Gulf. Southern Hope. Welborn Pet. Pollock. Texas Storm Proof. Williamson. Six Oaks. Truitt Premium. Zellner. Willis. The Arkansas Station1 classifies varieties according to habit of growth, placing them in two groups — the "long limb," which is charac- terized by its long and spreading branches, want of prolificacy, large bolls, late maturity, and long staple; and the " short limb," character- ized by its short branches, medium or small bolls, prolificacy, early maturity, and short staple. The same article also says that the varie- ties of the "long-limb" group grow slowly, require less readily available plant food and less frequent cultivation than the "short limb," which is regarded as being a direct product of high culture and especially '■ Arkansas Sta. Rpt. 1893. 218 THE COTTON PLANT. suitable for high farming. It also states that the short-limbed and cluster cottons can not be made to produce as long a fiber as the long- limbed varieties, and suggests that in many cases length of staple should be sacrificed for quantity, or more strictly speaking, that it is often better to sacrifice length of staple in order to secure an early matured and increased crop winch may be picked at a low cost, and before it is damaged by winds and rain. ' RELATIVE VALUES OF VARIETIES. In deciding upon the relative values of different varieties of cotton the planter will be guided wholly by the number of dollars per acre which each variety will bring, and what variety he shall choose is a question to which no definite reply can be given. The same variety will give different yields in different years, on account of early or late frosts and the amount of rainfall in different months, especially from August to October. Some varieties are more liable to suffer from insects than are others. Some varieties which make a long, fine, and silky staple when grown on rich, damp, and alluvial soil, such as is found only along the larger rivers, are of but little value when planted on the high and dry uplands. Some varieties respond much more quickly than others to applications of fertilizers; some produce a moderate crop under almost any conditions, while others give an enormous yield when ail the surroundings are favorable but fail miserably when planted on unfavorable soils or with unfavorable climatic conditions. The rela- tive prices of the lint from different varieties vary greatly from one year to another. In 1887 the lint from the long-staple sorts, those having a staple 35 mm. or more in length, sold in the New Orleans markets for nearly or quite double the prices paid for the short staples, or those having a staple 20 to 30 mm. in length. Since then, however, the prices have been approaching each other more and more closely, until now the difference is only from 10 to 20 per cent. The yield in pounds per acre for the short-staple varieties is always larger than for the long-staple sorts, though the amount of this difference will vary greatly with the localities where the crops are grown. In the hill regions the yield of the short-staple varieties is always much the greater, and is often fully double that of the long staples, and even on the low- lands of the southern portion of the cotton belt, the region best suited to the growth of the long staples, there is always a difference of yield in favor of the short staples. The long-staple varieties need not only a rich soil but the best of care in cultivation, handling, and ginning, and require the highest intelligence and skill for their proper management. Although some of the short-staple varieties are late in maturing their crop, still there are varieties which do mature much earlier than any of the long-staple varieties. In the southern portions of the cotton-grow- ing region, where the season is long, this is a matter of little moment, but it becomes a point of the greatest importance in more northern latitudes, where the season without frost is not sufficiently long to CULTIVATED VARIETIES OF COTTON. 219 mature any but the earliest ripening* sorts. No variety of cotton will be profitable to the planter if it does not have a sufficiently long- grow- ing season in which to mature its entire crop, and it is safe to conclude that the short-staple varieties AYill usually be found the more profita- ble in that part of the cotton region north of latitude 32°. The character of every plant is fixed in the seed from which it comes, and the peculiar vital force in each seed is determined to a great degree by the conditions surrounding the parent plant. Heredity plays an important part, but is not so strong as to wholly overcome the modifying influences of climate, soil, and season when these are acting in the same direction through generations. The law of the survival of the fittest applies to cotton as elsewhere, and when assisted by intelligence will soon develop iu each region the special type of plant best suited to that particular locality. A reference to the descriptions of the long sta- ple varieties will show that of the 17 varieties named in group G (p. 217), 1 is probably a hybrid with Sea Island, and that the histories of 2 otbers are unknown. Of the remaining 14, 6 originated in Mississippi, 3 in the adjoining State of Louisiana, and 3 others were developed in Georgia from seed sent there from Mississippi, so that the low alluvial soils and moist climate of the Mississippi River region have furnished the nec- essary conditions for the development of 12 out of the 14. In no other part of the cotton-growing region are the long-staple varieties grown so largely, and nowhere else do they approach so closely to the yields made by the short-staple sorts. Similar conditions exist along a few of the larger rivers in other sections, and in such locations the long- staple sorts will be equally successful, but such areas are quite limited in extent. Along growing season, a rich soil, and a humid atmosphere are all essential to their success, and it is useless to plant them where all of these conditions do not exist. The great increase in the amount of Sea Island cotton grown in this country, together with the rapidly increasing exportation of long staple cotton from Egypt, will do much to keep down the prices of the upland long-staple varieties, and it is doubtful whether they will ever again be as profitable as formerly for general cultivation. North of latitude 32° the growing season is so short that only those varieties which make their entire growth quickly can be grown with profit. Among the earliest ripening varieties are Cherry Cluster, Dickson, Drake Cluster, Hawkins, Hunnicutt, King, Peerless, and Peterkin. Of these 8 varieties 6 belong to group 4, having a staple less than 25 mm. in length, and in only 2 of them — King and Peerless — do we find representatives of group 5, with a staple from 25 to 30 mm. All belong to group 2, having from 30 to 34 per cent of lint. We may therefore safely conclude that the best varieties for the northern part of the cotton region will be such as have a short or medium length of staple, together with a fair but not excessive percentage of lint. The long-staple sorts and those producing more than 34 per cent of lint mature too late to be grown with safety in this region. 220 THE COTTON PLANT. In the middle and southern portions of the cotton belt early maturity is less important, though often an advantage by enabling the planter to gather the crop without injury from rains, which are often heavy and frequent in November and December. There the principal point to be considered is which variety will yield the greatest number of pounds of lint per acre, as there is very little difference between the selling prices of 20 mm. and 30 mm. staples. The common varieties which may be relied upon to produce 84 per cent or more of lint are Bates Big Boll, Brannon, Catawba, Excelsior, Jenkins, Peterkin, Bio Grande, and Texas Storm Proof. Of these 8 varieties G belong in group 4, those having a staple of less than 25 mm., and the other 2, each of which produces only 33 to 35 per cent, belong in group 5. An excessive percentage of lint and a long or even a medium length of staple are never found in the same variety. The difference in yield per acre between the short and the medium staple varieties is not marked, though there appears to be a slight difference in favor of the shorter staples. We have records of 881 tests of 59 varieties which have been made at the Southern experiment stations during the last seven years, each variety having been tested three or more times. By taking the average yield of all the varieties tested at each station during each year as 100, the relative yield and rank of the different varieties is found to be as follows : Relative rank as regards yield of different varieties. [The average for all varieties taken as 100. J Variety. Haggerman Texas Wood Taylor Brannon Peterkin Eishburn Thomas Kieth Drake Cluster Excelsior King Boyid Proline Truitt Premium ... Barnett Duncan Hunnicutt Jones Improved Peerless Texas Storm Proof. Early Carolina Bailey Kogeirs Dickson Deering Peterkin Cluster. . . Shine Early Welborn Pet Ben Smith. Dean Crawford Peerless . O Mi a* P S P a V, a a ■3 0 3 3 149 101" 131 9 143 105 123 7 177 06 119 9 140 102 117 53 171 70 116 4 149 95 116 5 157 101 115 - 5 134 100 114 6 152 88 112 10 154 84 109 44 173 76 108 10 156 68 107 45 161 40 106 6 126 83 100 8 129 86 105 8 134 74 105 23 135 82 105 30 143 76 105 27 176 69 104 7 126 83 104 4 132 79 104 3 122 94 104 21 141 74 103 19 150 80 103 5 124 80 103 18 132 61 103 37 185 73 103 11 139 73 102 3 105 99 102 20 155 58 101 Variety. Hawkins Ozier Southern Hope Ellsworth Oats Cochran Eureka Chambers Ethridge Hays China Jowers Willis Cherry Long Staple . Jones Long Staple . . . Peeler Petit Gulf Okra Allen Bolivar County Barneses Zellner Cobweb East Cherry Cluster Matthews .. Colthorp Pride Cook Six Oaks Wonderful tw ti a a 11 a s & tt N 34 148 57 17 119 45 34 139 68 8 125 54 7 120 05 6 130 76 9 140 78 3 101 89 3 117 68 5 114 84 11 128 76 3 112 72 14 132 74 26 136 57 8 108 81 15 132 46 28 124 67 34 144 52 3 116 83 6 105 83 9 105 44 5 103 87 3 113 75 13 123 55 6 111 62 3 99 71 12 123 56 4 126 57 4 113 44 101 100 98 98 97 97 97 97 97 00 96 06 96 95 94 91 94 04 93 92 90 8S 87 85 82 yo CULTIVATED VARIETIES OF COTTON. 221 It should be remembered that the table above refers to the relative yield of lint only, and does not bear on the value of the lint. The figures have been compiled from the results secured at 8 different experiment stations during seven years, and show the wide variations which may occur with any variety when grown in different localities and in different seasons, and illustrate very forcibly the statement made on a previous page that no one variety can be regarded as being the best for all soils and for all seasons. Of the varieties which have been tested 25 or more times, so that their relative values may be regarded as being fairly well known, Peterkin, King, and Truitt Pre- mium take the highest rank, followed closely by Peerless and Texas Storm Proof. Of these, Peterkin, Texas Storm Proof, and Truitt have a staple of less than 25 mm., while King and Peerless have a staple of from 25 to 30 mm. King, Peerless, and Truitt yield from 30 to 34 per cent of lint, and the other two — Peterkin and Texas Storm Proof — yield more than 34 per cent. It is also significant that the 5 varieties at the bottom of the list are all long-staple sorts, which produce less than 30 per cent of lint. As the table has been made up from the results secured at stations in both the northern and the southern parts of the cotton belt, it is possibly misleading in that it does not show what varieties have given the best results at either extreme, but it probably indicates very closely the relative standing of the different varieties in the central belt. Of the 5 varieties named 2 are early, 1 is medium, and 2 are late in time of maturity. All are "short" or "medium" staple, but it should be noted that there is no experiment station located in what is known as the "Delta" country along the Mississippi Eiver, the region which is universally recognized as being best adapted to the growth of the long-staple varieties. As showing how yields may vary, even in a single locality, the fol- lowing table is given from the results secured in the work of five years at the South Carolina Station, where the same varieties were used continuously; the figures indicate the relative rank as regards yield of each variety : 1 Relative rank as regards yield of different varieties. Variety. Peterkin Jones Unproved .. Dickson Improved Cobweb Thomas New Texas Ozier Common Duncan Richardson Dickson Cluster.. Hays China Crawford Peerless Average. 1 South Carolina Sta. Rpt. 1888, p. 218. 222 THE COTTON PLANT. These variations are nearly or quite as great as when the cottons are grown in different States, as may be seen by the following table, show- ing the relative rank as regards yield of eleven varieties, grown in four different States as reported by the Arkansas Station:1 Iielative rank as regards yield of different varieties. Variety. Peerless Shine Early Mammoth Prolific Hawkins Deering Herlong Allen Jones Long Staple Peterkin Southern Hope . . . Truitt Premium . . Arkansas. Georgia. Louisiana. 1 5 3 2 3 4 8 6 2 4 5 6 4 5 7 7 6 8 9 2 9 1 1 10 11 8 3 7 Mississippi. The Alabama College Station reports2 as a result of examinations of 25 varieties that the strongest cotton fibers were produced by Truitt, the largest by Barnet, the smallest by No. 1 (Peerless), Hawkins Improved, and Peterkin, the longest by Okra Leaf, the shortest by No. 2 (Peerless), and the best twisted by Truitt, Eameses, and Cherry Cluster ; the largest j>ercentage of fiber per boll was produced by Wel- born Pet, Okra Leaf, Peterkin, Hawkins Improved, ami King Improved, in the order named ; the largest percentage of seed per boll was pro- duced by Zellner, Eameses, Southern Hope, and Truitt, in the order named; and the best grade of cotton, taking all things into considera- tion, is Cherry Cluster, the second best Truitt. Alabama Canebrake Station, in reporting a test of 13 varieties, says:3 "Peerless and Welborn were the best types of cluster cotton. The greatest yield was made by Peerless, Peterkin, and Texas Storm and Drought Proof." The same station, in reporting tests of 15 varieties in 1891, says: "Peterkin and Peerless are the most desirable varieties." The directions in which we must look for a further improvement in varieties will vary somewhat with the latitude where the crop is to be grown. In order to be desirable for any locality the crop must mature before the plants are killed by frost, so that, at least for the northern cotton belt, it is essential that complete and early maturity be secured. No variety which does not mature fully three-fourths of its crop before the first of October should be grown north of latitude 32°, and even south of that region early maturity is desirable as enabling the crop to escape probable injury and loss from October storms, though it is not as essential as farther north. Individual plants of any variety differ from each other so widely that it is an easy matter to select certain ones which have characteristics 1 Arkansas Sta. Bui. 18. 2 Alabama College Sta. Bui. 13 (11. ser.). 3 Alabama Canebrake Sta. Buls. 11 and 14. CULTIVATED VARIETIES OF COTTON. 223 making- them more desirable than are the majority of those in the same field. An ideal plant should be vigorous in its growth and wholly free from disease. The branches should be sufficiently strong and rigid, so that when they are loaded with green bolls they will not touch the ground. There is a great diversity of opinion as to the relative pro- ductiveness of the "cluster" and the "limbed" varieties, both styles of growth having their warm supporters. Many planters claim that the cluster or short-limbed varieties are best on deep and rich soil, while the longer-limbed sorts are superior for lighter soils. The results which have been secured in a large number of comparative tests, however, do not seem to show any foundation for this opinion, and indicate that the differences, if any, are attributable to other reasons than the mere length of the limbs. The short-limbed and cluster sorts can be planted more closely than can those varieties having longer and more spread- ing branches, but usually have smaller bolls, and none of the cluster varieties produce a long staple. Many of the cluster varieties have sharp points to the bolls, which make them unpleasant to pick, though this objection does not apply to all. Most of the cluster varieties mature early, though none of them mature as early as do some of the medium and short-limbed sorts, and a few mature quite late. The general results of the tests made show the cluster varieties to be better adapted for cultivation in the middle and southern cotton region than ill the extreme northern belt. The bolls should be of good size, and a cross section should show a circle and not a triangle with rounded corners. When mature they should point downward rather than upward, so that rain and heavy dews will not enter and rot the con- tents; and they should also open widely enough to permit easy pick- ing, but not enough to allow the seed cotton to drop to the ground. It is of no advantage to have more than three or four divisions or "locks" in a boll, as when spherical in shape, and with a given diameter, its contents can be no greater in six sections than in three. The bracts at the base of the boll should be small, so that they will not be in the way and become entangled with the seed cotton in picking. The lint should be pure white in color, strong, even, silky, uniform in length and twist, and should be easily separated from the seed in ginning. The staple should be not less than 25 mm. in length, and if 30 mm. can be secured it will be still better, though it will not be advisable to sacrifice a larger yield for a longer staple so long as the present rela- tive prices of the long and short staples are maintained. With early maturity and a medium length of staple secured, the only additional feature which need be considered is how to secure the greatest possible yield of lint per acre. It is often claimed that the larger the percentage of lint to seed cotton the more valuable the variety. This belief has no foundation in fact, and is really the reverse of true, provided the total yield of liut per acre remains the same. A crop of 500 pounds of lint and 1,100 pounds of seed per acre is certainly more 224 THE COTTON PLANT. valuable than the same crop of lint with only 900 pounds of seed. At present prices the seed is an important part of the crop, and if the yield can be increased without detriment to the total yield of lint it will add just that much to the value of the entire crop. In speaking of this matter the Georgia Station says : ' It maybe said by way of caution tbat tberc is no necessary relation between the yield of lint per 100 pounds of seed cotton and the actual yield of lint per acre. A variety may yield a high percentage of lint, calculated on a giveu weight of seed cotton, and yet yield less lint per acre than another variety. It is probably more a question of seed than of lint. As the seed contains nearly all of the valuable ele- ments taken from the soil, it is but reasonable to expect that a large yield of seed per acre will be attended by a correspondingly large yield of lint. We have but little doubt that some varieties that are popular with the mass of farmers because of their percentage of lint compared to seed simply produce less seed per acre instead of more lint per acre. It remains true, however, that a small-seeded variety — small in size and small in percentage of the whole — is better for poor land and low culture than a A^ariety having a naturally large seed and a smaller percentage of lint. It is a well-established fact that large seeds will produce stronger and more vigorous plants than small seeds of the same variety. Cotton is no exception to the rule, and if bred to produce small seeds it will be at the expense of the size, vigor, and health of the plant. A heavy yield of good lint is the main object to be attained, but this will not necessarily follow from a decrease in the size of the seed. Of course, the planter can not afford to increase the percentage of seed at the expense of the lint, but if the pounds of lint per acre can be maintained the greater the amount of seed the greater will be the profit. Whether or not the weight of seed can safely be increased without a decrease in the yield of lint is a matter which has received almost no attention, but which is a promising field for the investigator. BIBLIOGRAPHY. Tenth Census of the United States, vols. 5 and 6. Cotton Production, parts 1 and 2. Cotton Planter's Manual, Lyman. Dictionary of Economic Plants, Watt. Alabama College Sta. Bui. 4, 1888; Bui. 5, n. s., 1889; Buls. 12, 16, n. s., 1890; Buls.22, 23,n.s., 1891; Bui. 52, 1894; Alabama Canebrake Bui. 7, 1890; Bui. 11, 1891; Bui. 14, 1892. Arkansas Sta. Bui. 18; Repts. 1888, 1889, 1890. GeorgiaSta.Bul.il; Bui. 16; Bui. 20; Bui. 24. Louisiana Stas. Bui. 13; Buls. 21, 22; Buls. 26, 27; Buls. 7, 8, 2d ser.; Buls. 16, 17, 2d ser. ; Buls. 21, 22, 2d ser. ; Buls. 28, 29, 2d ser. Mississippi Sta. Bui. 18; Bui. 23; Rpts. 1889, 1890, 1891, 1892, 1893. North Carolina Sta. Rpt. 1887. South Carolina Sta. Bui. 1, n. s. ; Bui. 2, n. s. ; Rpts. 1888, 1889. Texas Sta. Rept. 1889. 1 Georgia Sta. Bui. 20. CULTURE OF COTTON. By Harry Hammond. GEOGRAPHY OF THE COTTON BELT. The cotton belt covers 24° of longitude and 10° of latitude. Exclud- ing from the count the greater part of Virginia, more than 100,000 square miles of western Texas, and the whole of Kentucky, Kansas, Missouri, Utah, California, Arizona, and New Mexico, in all of which cotton has been cultivated and where a larger demand might cause its culture to be extended, the cotton-growing region measures nearly 600,000 square miles, almost one-third of the total area of settlement, in 1890, of the United States. The 20,000,000 acres planted in cotton occupies barely 5 acres in every 100 of this extensive region. Scarcely 50 per cent of this territory is in farms, and not more than one-fifth has at any time been tilled. This section contained in 1890 a population of over 8,000,000 whites and something over 5,000,000 negroes, in all 13,651,000, every 100 of them producing 53 bales of cotton, an average of 254 pounds of lint per capita. The Mississippi Eiver, turned from its southeasterly course to one south of west by the bluff lauds of Tennessee, Mississippi, and Louisi- ana, divides the cotton belt into two nearly equal eastern and western portions. Bordering the west of the flood plain of the great river are the oak and hickory uplands of Arkansas, Louisiana, and Texas, stretching westward more than 200 miles to the black Cretaceous prai- ries of Texas. These black prairies descend from Indian Territory in a broad crescent, its concave edge facing west and inclosing the more elevated red-loam prairies until it thins out into the coast prairies of the southwest in the neighborhood of Austin. To the north the oak and hickory is bounded by the red lands of Arkansas. The counter- parts of these regions are found east of the Mississippi. Moving east from the bluff and yellow-loam table-lands that rise from the flood plain of the river, we again meet the oak and hickory lands in Mississippi and Alabama. Beyond these another crescent of black Cretaceous prai- ries, its concave edge, however, facing east, reaches from northwestern Mississippi to southeastern Alabama. Northeast from the concave border of these prairies lie the valley lands of Alabama, Georgia, and Tennessee, the Coal Measures of those States, the gravelly hills of Ala- bama, and the central basin of Tennessee. Near the termination of these prairies in southeastern Alabama another region is encountered. 1993— No. 33 15 225 226 THE COTTON PLANT. This is a prolongation of the Alleghanies, and passes from the local- ity named, in a broad belt, to the northeast, across the States of Ala- bama, Georgia, South Carolina and North Carolina. It is known as the metamorphic or Piedmont region, or, in popular parlance, as the region of granitic rocks. On the northeastern border of this region, in North Carolina, the pine hills are met. The pine-hills region reaches south- westward along the southern border of the Piedmont region and the Alabama and Mississippi prairies, traversing all the Atlantic and Gulf States, interrupted only by the delta of the Mississippi, until it crosses Louisiana and reaches the oak and hickory of Texas. South of the pine-hills region is a broad belt of level pine lands, coextensive with it and everywhere touching either the Atlantic or the Gulf coasts, until it reaches the coast prairies of Louisiana and Texas, where it terminates. In 1801 South Carolina led the other States in the production of cot- ton. In 1850 Alabama stood first in the amount produced. Mississippi led in 1860-1880, and Texas stood first in this respect in 1890. CENTERS OF COTTON PRODUCTION. The center of cotton production, the point where the weight of all the cotton produced in the various regions of the cotton belt would stand in equilibrium, was located in 1850 a few miles north of Montgomery, Ala. In 1860 this center had moved 200 miles west to a point some 20 miles northeast from Jackson, Miss. Its movement in 1870 was north- east to a point near Carthage, Leake County, Miss. It again moved northeastward into Noxubee County, Miss., in 1880. Its movement was about GO miles northwest in 1890 to Kosciusko, in Attala County, in the same State. It is probable that since that date the increasing crops of Arkansas, northern Texas, and Indian Territory have drawn the center of cotton production still farther to the northwest. THE PINE LEVELS. These extend inland from the Atlantic and Gulf coasts for from 50 to 150 miles and reach an elevation of some 200 feet above sea level, embracing an area of 34,000,000 acres. Forty- four per cent of this area is in farms. Eight per cent was improved land in 1890, being an increase from 6 per cent in 1880. Only 1 per cent of these lands was planted in cotton iu 1880, but double that amount was put in in 1890. They pro- duced 2.G and 3.2 per cent of the total cotton crop at the beginning and close of the eleventh decade, respectively. The cotton acreage was increased during this period 57 per cent and the cotton crop 62 per cent. This was of exceptional occurrence, for it almost always happens that an increase in acreage is not accompanied by a proportional increase in the yield. Both these increments were considerably greater than the increase of the population, which was only 22 per cent. Nevertheless, the pine levels are not given as exclusively to cotton growing as most of TJ. S. Dept. of Agriculture, Expt. Sta. Bui. 33. Plate III. CULTURE OF COTTON. 227 the cotton belt is ; for while the average product per capita of the cotton belt was 231 pounds of lint cotton in 1880 and 259 in 1890, only 85 pounds of lint per capita was produced in 1880 by the inhabitants of the pine levels, increasing in 1890 to 99 pounds. This was not due to lack of fertility in the soil, for the product per acre was equal to, and in instances greater, than some of the other cotton regions, and it only required in 1890 two-tenths of an acre more than the average of the cotton belt to produce a bale of cotton here, which, considering the very easy tillage of these light sandy loams m comparison with most others, did not appreciably increase the cost of production. The yield was a bale to 3.1 acres in 1880 and a bale to 2,9 acres in 1890. A white population preponderates in 61 of the 89 counties consti- tuting this region. The only State in which the negro population is in excess is South Carolina, where they were settled long ago upon the fertile rice fields and the sea islands, producing long staple silk cotton. The colored population was only 19 per cent of the population in this region in 1880 and 25 per cent in 1890. That 75 per cent of the popula- tion here is white is an observation running counter to the generally received opinion that the moist, warm climate of the low latitudes is unfavorable to the white man, and that nature intended them for the black race. An observation of a similar character in regard to the work animals employed in agriculture will have to be changed. The mule has been thought especially fitted for work in low latitudes, and the South generally has maintained this barren stock at heavy cost. Here, in the lowest latitude, the number of horses exceeds that of mules in 76 out of the 89 counties in the pine levels and coast region, even where the colored race preponderates, they being supposed to prefer the mule. The farms here are larger than anywhere else in the cotton belt, averaging 233 acres to the farm in 1880, and 190 acres in 1890, against a general average for the cotton belt of 119 acres to the farm. Sixty-nine per cent of these farms are occupied and worked by their owners, only 31 per cent being rented, in which regard this region again offers a striking contrast to the remainder of the cotton belt. The tillage of these lands is easy. The average is 37 acres of improved land to the work animal as against 22 for the whole cotton belt. Three bales for each work animal is the average, which is also high, the average for the cotton belt being 2.1 bales. The great agricultural need of the pine levels is drainage, and this, under the increasing subdivisions of the farms, has not received proper attention, and in fact it has not been practicable to drain the land properly. Owing to their immense areas of swamp land, their rich vegetable mold, resting on marl or phosphate rock, has remained untouched. The general culture of cotton is otherwise essentially the same as that to be described in the pine-hills region. There are, however, two notable exceptions — the culture of Sea Island cotton in 228 THE COTTON PLANT. South Carolina and Georgia, and of the same cotton in the interior of Georgia — which should have special mention. Culture of Sea Island cotton. — On the sea islands of South Carolina field labor is performed almost exclusively by negroes. Xearly all of them are engaged in farming on their own account; a large number own farms; a still larger number rent lands for cultivation, and even the laborers are paid most generally by granting them the use of so many acres of land for certain stipulated services. The largest number of acres of Sea Island cotton planted under one management scarcely anywhere exceeds 100 acres. The white planters do not average prob- ably more than 30 acres, and this requires that they should be land- lords of considerable estates; for, as the laborers are frequently given 5 to 7 acres for two days' work in the week, and as two days' work per week does not suffice for the cultivation of more than 4 acres, to culti- vate 30 acres of cotton under this system requires in addition 75 acres of land; add to this the amount usually planted in corn and other crops, and we will have 120 acres. As, under the best system, the land lies fallow every other year, the planter of 30 acres of cotton will require 210 acres of open land; and as scarcely one-fifth of the land is under cultivation, such a planter will probably own 1,200 acres. This state of things is owing to the scarcity of capital, and to the low price of land and labor. Drainage, although said by Governor Seabrooke to be so little attended to, has of necessity always been practiced to some extent on the sea islands. The remarkably high beds on which the cotton is planted here — 18 inches to 2 feet high — subserve this purpose. The best planters have long had open drains in their fields. These were gener- ally made by running two furrows with a plow and afterwards hauling the loose dirt out with a hoe, thus leaving an open ditch, if it may be so termed, a foot or more in depth. In recent years the farmers of James Island have made deeper ditches and placed plank drains in them. Seeing the great benefit resulting from this, they subsequently replaced the plank with regular drainage tile. The outlets open on the sea at the low- water mark, and the pressure of the water on the pipes preserves a constant outflow, even at high tide. In this manner, land only 1 or 2 feet above high water is susceptible of thorough drainage to the depth of 4 and even of 5 feet. The borders of these islands being usually the highest part and the richest land in the interior often much lower, a wide field for improvement is offered in this direction. In the early part of the century, when agriculture had so far devel- oped the value of these lands as to make $60 an acre a not unusual price, the use of the plow was entirely unknown here and all the opera- tions of tdlage were performed by hand with the hoe alone. This con- tinued to be the usual practice until 1865. Since then plows have come more and more into use, until their employment became quite general. Fallowing is practiced to the extent that land planted in cotton one CULTURE OF COTTON. 229 year is pastured the next by cattle and sheep but not by hogs, and it is claimed that great benefit is derived by having the loose soil of these islands thus trodden by stock during the year they lie fallow. The rapid growth of bushes, briers, and weeds is kept down by the stock, and the large growth of cotton stalks of the previous year, after fur- nishing food for the animals, is broken up and scattered. If care be taken "that the grass is not eaten so close as to expose the soil on the top of the beds to the summer sun," it is found when the stock is turned off in November to range through the cultivated fields that the pasture uis in exactly the right condition for the coming season's cotton fields, with no cotton stalks or troublesome growth to be got off or under the land and make it too husky." A mule can do the plowing required in the cultivation of 30 acres of Sea Island cotton, and can, in addition, cultivate a sufficiency of land to supply corn for its own feed. The first step in the preparation of the land is to hoe off the weeds ("hurricane"), cut up the cotton stalks, and pile and burn this litter. This costs 40 cents an acre. Bushes are grubbed up at 7 cents an acre. The land is not thoroughly plowed, but in February two furrows are run with a single-horse turning plow in the middles between the old beds, leaving a trench 7 or 8 inches deep. In this furrow a subsoil plow may or may not be run, according to the character of the soil — whether wet or dry. On James Island, where underdrainage is practiced, this furrow is generally used. Before plows came into use this trench was never made, and even now it is omitted by some of the most successful planters. In this trench, or in the middle of the alley when no trench is made, the manure is placed. This consists usually of 20 cart loads of marsh mud and 1,000 to 1,400 pounds of cottou seed to the acre. Stable and lot manure, together with compost of marsh mud, are also applied at the rate of 40 cart loads to the acre on such portion of the land as the lim- ited number of stock belonging to the farmer enables him to treat in this manner. On the lines of manure thus laid down a certain quantity of commercial fertilizers is drilled. This practice, formerly unknown, is very common now, even the smallest negro farmers often going heavily in debt to obtain these fertilizers. The land is now ready for listing, which is done by hauling the soil from the top and sides of the old beds onto the manure with a hoe. A more recent practice is to lap in with two furrows of a turning plow on the manure. This last costs only 17£ cents per acre, while the listing with the hoe costs SO cents, although the latter has the advantage of bringing all the humus and vegetable mold directly to the spot where the roots of the plant are to grow. Over the mass of dirt, weeds, manure, etc., thus collected in the old alley a double roller 5 feet from center to center and weigh- ing about 800 pounds is passed to press together and compact the whole, completing two rows at a time. All this should be finished by the first to the middle of March, and the bed is then built up by lapping in two more furrows on a side with a single or double horse turning plow. 230 THE COTTON PLANT. The land is now ready for planting, which may begin any time after March 20, but April 1 to 10 is the time preferred. Cotton planters are not generally used. Three laborers do this work ; the one ahead chox)S a hole with a hoe, on the top of the bed, at intervals of 12 to IS inches; another drops eight or ten seed in each hole, and a third follows ami covers carefully with the hoe. Three to four pecks of seed are used to the acre. The seed makes its appearance above ground in eight to twelve days after it is planted, and the stand is perfected from the second week in April to the first week in May. Hoeing begins about the first of May. The second hoeing takes place the last of May. The plows then break out the middles (the spaces between the new beds where the old beds stood). The hoe hands follow and pull up the loose dirt left by the plows to the foot of the cotton. This is called u hauling,'' and by it the new bed is completed, the cotton is kept from "flagging" (falling down), and the grass is kept under. It costs 80 cents an acre. At the second hoeing some stalks are thinned from the bunch in which the seed breaks the ground, and at each succeeding hoeing and hauling other stalks are removed until in July only one stalk of each bunch is left. There are four hoeings and four haulings, one or more furrows with a sweep being run through the middles previous to each hauling. By the last of July the culture is completed, except to run a furrow between the rows in August to destroy the grass and keep the cotton growing. The first blossoms appear about the middle of June, when the cotton is 15 inches high, and the bolls open in August, when the plants have attained a growth of 4 to 5 feet. Cotton picking commences from the last week in August to the first week in September. By the 15th of December the crop is gathered. When the cotton has been picked, weighed, and housed, it is next spread out in the sun on what is called an " arbor." This is a platform 25 feet or more square made usually of inch boards. Here the sun and air dry the cotton, preventing it from heating, which it is liable to do when stored in bulk, and, it is also thought, causes the lint to absorb some of the oil in the seed, which adds to the silky luster of the fiber. After being thus dried, it may be either stored or passed "at once to the whipper, a machine that knocks the dust and sand out and leaves the cotton whiter and more open. Formerly it was all assorted. A hand was given 150 pounds of seed cotton as a day's task, which he thoroughly overhauled, picking out all specks, stained cotton, fragments of leaf, etc. At present, however, this is usually done by two hands, who examine the cotton as it passes to the gin, and two others behind the gin, who pick out cracked seed, motes, etc., as the lint issues from the gin. The roller gin in some form has always been and still is used for detaching the lint from this black seed cotton. The first successful crop of Sea Isfcind cotton was grown by William Elliott on Hilton Head in 1790. In 1S05 this quality of cotton sold for 30 cents per pound, while uplands were selling at 22 cents; in 1816 at CULTURE OF COTTON. 231 47 ceuts, with uplands at 27 cents. The crop of 1825 amounted to 2G,039 bales, of which 7,779 were grown in Georgia, and the balance in South Carolina. In that year Mr. Kinsey Burden, of South Carolina, sold GO bales at $1. 10 per pound. The same gentleman sold his crop of another year for $1.25 per pound, when the average price of uplands was 9^ cents. He also sold two bales in 1828 at $2 a pound, which is the highest price on record. The crop of Sea Island rose to 36,776 bales in 1829, of which 13,729 were made in Georgia, and the remainder on the sea islands of South Carolina. The crop fell off in 1839, South Carolina still leading with 9,975 bales, and Georgia making 1,225. The two States together made 20,184 bales in 1819, South Carolina making 18,921. During the fifties this culture was extended to Florida, and that State produced nearly 11,000 bales in 1856. In 1859 Florida led with a crop of 20,353 bales, while South Carolina made 13,391 and Georgia 10,352. The Sea Island crop exceeded 26,000 bales in 1869 and in 1879, in both years Florida producing a larger crop. In 1870 Texas contributed 704 bales of Sea Island cotton, and in 1871 made 1,100 bales; since that date the crop there has gradually dwindled down and there is no record of it after 1882-83. In 1889 the crop reached 16,841 bales, Florida producing 25,111 bales, Georgia 12,131 bales, and South Carolina 9,299. It was about this date, that this culture began to develop among the small farmers of the pine levels at a considerable distance inland from the sea. So successful has this adventure proved that in 1894 the Sea Island crop of Georgia rose to 39,367 bales, while that of Florida fell to 19,107, and South Carolina, where this culture had originated and flourished longest, fell back to a crop of 2,578. It had seemed for many years that islands on the South Carolina coast pos- sessed a natural monopoly for the production of the finest staple. At that time the culture was conducted under the superintendence of men of high intelligence, and the selection of the seed and the cultivation and preparation of the crop for market was attended to with great skill and the most scrupulous care. At present it is chiefly in the hands of small farmers of the colored race, whose intelligence, skill, and care are wholly occupied in securing a bare subsistence for themselves. It is doubtful if there is any local monopoly of the production of long- staple cotton. It has been grown successfully in the up country, more than 100 miles from the coast, and all the seed from which the finest strains of Sea Island cotton have been derived came from seed planted in the interior of South Carolina, for several years, during the late war. The extent to which the manufacture of these long staples in the United States has increased in recent years is noteworthy: In 1870 only 5 per cent of the crop was consumed in this country; by 1880 the consump- tion had risen to 35 per cent; and in 1894, although the crop had increased 130 per cent, nearly 40 per cent of it was consumed at home. In addition to that, a very large amount of long-staple cotton — perhaps more than 10,000 bales — was imported from Egypt for manufacture. 232 THE COTTON PLANT. The Egyptian cotton is inferior to the finest grades of Carolina Sea Island, and it is also inferior to the Santee and Florida, the growth of which has so greatly increased of late years. It competes with the long-staple uplands, the cultivation of which was very profitable some years since, but which have been almost entirely abandoned owing to being supplanted by imports from Egypt. The culture of Sea Island cotton in the interior of Georgia is not so elab- orate as the method described as pursued on the coast. Seed is brought every year from the coast in quantities to plant patches sufficient to fur- nish seed for the whole crop the ensuing year. The seed of the second season is thought to do better, but after that the staple deteriorates and seed is used which has been planted away from the coast only one year. They use a very effective homemade implement for cutting down the old stalks. It is a roller on which iron blades are fastened longitudi- nally. Shafts are attached and it is drawn by one horse along one row at a time, cutting the stalks into lengths of 8 to 10 inches. The general culture resembles that of upland cotton, and is usually performed entirely by the farmer and his family. They raise their own supplies and provisions to sell, such as poultry, eggs, honey, and country-cured hams. Elsewhere the falling off in the long-staple crops of the coast and Florida indicates a depression as great, or greater, in this culture as in any other branch of agriculture. The average value of farm lands in the pine levels is lower than else- where in the cotton belt, but it rose during the eleventh decade from $2.09 to $1.01 per acre, while unimproved land may be bought for 50 cents an acre. Five per cent of these lands in North Carolina are said to produce three-fourths of a bale to the acre without fertilizers, 20 per cent half a bale, and 50 per cent one-third of a bale. Five estimates of the cost of production there in 1880 ranged from 5£ cents per pound of lint to 10 cents, the average being 7.3 cents. Wharton gives the cost of the crop of 1891 as 7£ cents and of 1892 as 11 cents, owing to unfavorable seasons, and the average for three years as 8 cents. The crop of Mr. Nobles cost 7 cents in 1892 and 5£ cents in 1893, the reduced cost resulting from diminishing the acreage and planting other crops. In South Carolina the cost of growing Sea Island cotton was esti- mated in 1880 at from 15 cents to 21 cents per pound of lint, with a net profit per acre of $38 to $78 at the prices then prevailing. The cost of producing short staples in the interior was placed then at 6i cents; estimates in 1893 give the cost of short staple at 5 cents on the best and 10 to 11 cents per pound ou the poorest land. In Georgia the cost of producing short staple in the pine levels was estimated iu 1880 at 8 cents to 10 cents per pound; in 1892 it is placed at Ih, cents. The cost of growing Sea Island cotton was thought to be 50 cents a pound in 1880. Iu the Alabama pine levels about 1 per cent of the land produces, CULTURE OF COTTON. 233 without fertilizers, a bale to the acre, at a cost of 3 cents a pound ; 2 per- cent, three-fourths of a bale, costing- 3.1 cents; 15 per cent, half a bale, costing 1 cents; 30 per cent produces one-third of a bale, costing 5.8 cents. On the pine levels of Mississippi in 1880 cotton seems to have been produced cheaper than anywhere else in the State, the lowest estimate for the whole State, as to cost of production — 1 cents — being given there. In 1893 the estimate of cost of production is 1 cents on the best land and G^- cents on the poorer lands. THE PINE-HILLS REGION. This region stretches inland from the northern border of the pine levels along the whole extent of the latter, reaching an elevation of 200 to 400 feet above sea level. The country is rolling, and the open pine woods of the pine levels is replaced by a long-leaf pine woods, with an undergrowth of many varieties of oak, and some hickory. This region covers over 39,000,000 acres, 58 per cent of which is in farms. Twenty-two per cent of the whole area consists of improved land, and 7 per cent of it was planted in cotton in 1890. It produced in 1890 15 per cent of the whole cotton crop, the acreage having been increased 31 per cent, and the crop 37 per cent, during the eleventh decade. The increase in acreage was general except in the long-leaf pine hills of Texas, where a slight falling off was shown. It was greatest in South Carolina, amounting there to 01 per cent. The increase in the crop was not so uniform. There was an actual decrease in Xorth Caro- lina, notwithstanding the increased acreage, as well as in Texas. In South Carolina the increased acreage produced a larger crop by only 57 per cent, but in Georgia, Alabama, Mississippi, and Louisiana, the crops were increased very considerably more than the acreage. This region contains Burke, the county of largest cotton production in Georgia; Mecklenburg, the banner cotton county of Xorth Caro- lina, and Barbour, second in production among the counties of Alabama. The credit, however, of producing more cotton in 1890 thau any other county in this region belongs to Barnwell, in South Carolina, which grew 50,170 bales, placing it fourth iu the list of cotton-producing counties in the cotton belt for that year, the others being Abbeville, S. C, and Washington and Bolivar counties, Miss. The population did not increase so rapidly as the acreage and crop of cotton, being only 16 per cent greater in 1890 than it was in 18S0. Although the white population preponderates in G3 per cent of the counties constituting this region, being more numerous than the colored race in Xorth Carolina, Alabama, Mississippi, Louisiana, and Texas, nevertheless the pine hills of South Carolina and Georgia are so thickly settled with negroes that it made this race count for 51 per cent of the whole population of the region in 1890. The colored race was 53 per cent of the population in 1880. The per capita production 234 THE COTTON PLANT. of cotton amounted to 275 pounds of lint at the beginning and 337 pounds of lint at the close of the eleventh decade. The work animals are mostly mules, especially in Georgia and South Carolina, where there are large colored populations; in Mississippi, Louisiana, and Texas, where the whites preponderate, there are more horses. The acreage of improved land per work animal is 33 acres, an increase of 3 acres during the eleventh decade. This acreage is greatest in North and South Carolina (40 and 37 acres, respectively) than it is in Louisiana (19 acres) and in Texas (6 acres). Four and one-tenth bales of cotton were produced to the work animal in 1890, against 3.7 bales in 1880. In South Carolina 0.1 bales were made, in Georgia 6.1, and in Alabama 4.8, and less in the other States, until Texas oidy produced 0.9 of a bale. But everywhere with the increase of arable land to the work animal there was also an increase in the cotton pro- duced to the animal. The average size of the farms in the pine hills in 1 890 was 134 acres, being a decrease of 27 acres to the farm since the census of 1880. Forty- three per cent of these farms were under 50 acres, an increase in the number of farms of this class of 3 per cent during the decade. During this period there was a decline in the number of farms worked by owners of 5 per cent, only 53 per cent belonging to that class. There was no increase iu the number of farms rented for a share of the crop, and the larger number which were rented were rented for fixed money rents. Marlboro County, in South Carolina, is a typical cotton county of this region. It produced in 1890 32,306 bales of cotton on 58,836 acres, being a little over a bale to 1.8 acres, a yield not exceeded anywhere except in the alluvium of the Mississippi Kiver. The prevailing tendency of agriculture in the cotton belt may be illustrated by the changes in the expenditure of agricultural energy that have taken place in Marlboro in the last fifty years, exhibited by the per capita production of the leading staples of the farm, as follows: Production of crops and farm stock per capita in Marlboro County, S. C. Tear. Cotton (lint). Cereals. Horses and mules. Keat cattle. Swine. Sheep. 1840 Poxinds. 71 437 647 Bushels. 36 29 21 0.20 .20 .16 1.10 .66 .11 1.90 1.68 .39 0.34 I860 .28 1890 .01 The change has been continuous and progressive from a more or less mixed husbandry to one which has made everything subsidiary to the production of cotton. The lauds of Marlboro were thought to be exhausted in the early part of this century, and numbers of the popu- lation emigrated to the fresh lands of Alabama. The lands they left were level and wet. An extensive system of drainage was instituted, and in the course of forty years it is thought that the general water CULTURE OF COTTON. 235 level has been lowered 5 feet. This was the foundation on which they built. From one-third in some parts to two-thirds in others of the field work was done by whites in 1880, notwithstanding' that the colored population Las largely increased since the improvement iu the cotton crop, their services being iu requisition for picking that crop. The size of the farms is larger than in the other counties of this region in South Carolina, and the occupancy by owners is greater. Very little of the land lies fallow. Cropping is continuous. Fields planted in cotton for fourteen successive years produce better than they did at first. Eota- tion of crops when practiced is similar to what it is elsewhere — cotton one year, corn the next, followed by oats in the fall, the oats followed the ensuing summer by corn and peas, or by peas alone. The fourth year the stubble is broken up, 8 to 10 inches deep, and cotton planted again. The culture of very little land has been abandoned of late. Within the eleventh decade the improved land has been increased 27 per cent, and the estimated value of farms 74 per cent. Green manuring with the cowpea sown broadcast has been exten- sively practiced, and when cotton is laid by, peas are often drilled between the rows, where the beds for the next year's cotton crop is to be thrown up. All of the cotton seed, or its equivalent in meal, is returned to the soil, either alone or composted with stable manure, woods mold, and superphosphate of lime. In 1880 an average of $4.77 per acre for each acre in cotton was expended in commercial fertilizers. After clearing the old stalks, or breaking the stubble land, a furrow with a shovel plow was run in the old alley, or the land broken broad- cast was laid off in 4-foot rows with the same implement. The cotton- seed compost and stable manure were placed in this furrow and the bed built upon it with single-horse turn plows. This should be done early in February to prevent the cotton seed from sprouting and coming up. Another point is to get the manure covered as deeply as practica- ble, so as to keep it moist. Even after the greatest care, the farmer is often disappointed to find his manure brought to the surface during cultivation, where, exposed to the dry heat of summer, it is of no avail. It appears that the particles of the soil are not at rest, but under various influences, especially rains, they cave in and settle. The law of specific gravity operates, and in time the diverse components are assorted and find their respective levels as surely as cork floats or lead sinks in water. Planting takes place in April, usually with a seed sower. Early planting is best, except for the risk of damage by frost, for late plant- ing may lack moisture to come up to a stand. The after culture is with the hoe, to keep the cotton clean in the drill. This is most easily done in fields planted continuously in cotton, because the long summer cul- tivation, especially if done thoroughly late in the season, is very effect- ive in destroying the seeds of every variety of grass and weeds. Each 236 THE COTTON PLANT. hoeing is followed by the plow, whicli throws the dirt close np to the stalk. The cultivation of cotton throughout the pine hills does not vary much from this. In some places the labor of knocking down the cotton stalks with a club or pulling them is avoided by using a stalk cutter. The usual form is a gum log 18 inches in diameter and G feet long set to roll in a frame, to which a tongue for two horses is fitted. Bars of iron 3 inches by 4- inch, sharpened on one edge, are fastened edgewise by staples to the log. A man and two horses will cut up the stalks on 10 to 12 acres a day with this implement, taking two rows at a time. Before the profuse use of commercial fertilizers was depended on so largely, the cotton growers considered it of extreme importance to have the plants exactly spaced, so that each might; obtain its share of what the unaided soil had to offer. This was effected in various ways. A man was furnished with a dibble to make holes, in which he dropped the seed, covering them with his foot. A pointer extending from the dibble marked the spot where each successive hole was to be made. Another arrangement was to drag the beds off, two at a time, with a drag pulled by a horse walking in the alley. This left a fresh surface free of grass. The drag was followed by a horse pulling two wheels, on which cogs, spaced at the desired distances, made the holes for the seed in the two rows that had been dragged off. The dibble was followed by four women or children, who dropped the seed in the holes, and the operation was completed by covering a single row with a scraper pushed by hand, or two by a board drawn by a horse. Nearly perfect stands were obtained and kept by this precision, for there could never be any doubt when a hill was missing or was cut up by the hoe hand, and careful replanting was required. There were 210 hills to the acre row, 60 rows to the acre ; and if the stalks bore 6 bolls on the average, half a bale was made to the acre, a fair and pretty certain average for land with- out commercial fertilizers, depending for restoration on rest each alter- nate year and such stable manure as could be obtained. Now 6 bolls to the stalk would be a failure, for the spacing of the hills being left entirely to the judgment of the hoe hand, they stand anywhere from 1 to 3 feet apart, the deficiency in their number being corrected by the application of commercial fertilizers until it has become an axiom that cotton will produce equally well when planted 3 by 3 feet, or 4 by 4 feet, or 4 by 1 foot. Shallow culture has always been aimed at in the light, loamy soil of the pine hills. Notwithstanding more or less depressing reports, owing to the increase of the population and to the increments in repairs and improvements which convert the daily labor of the farmer into the fixed capital of the country, farm values have risen in the pine hills region during the eleventh decade 42 per cent. The largest increase (152 per cent) took place in Louisiana and the least (24 per cent) in North Carolina. The CULTURE OF COTTON. 237 other States stood in the following order: Alabama, 70 per cent; South Carolina, 01 ; Mississippi, 59; Texas, 30, and Georgia, 29 per cent. The increase in the value of agricultural products for the whole region was 22 per cent, ranging from an increase of 98 per cent in Louisiana to a decrease of 17 per cent in the pine hills of North Carolina. The other States stood in the following order: Alabama, Mississippi, Georgia, South Carolina, Texas. The average value per acre of farm lauds was $0.18 in 1890, an increase from $4.54 in 1880. These values were greater in the eastern and older States, and less in the western and younger ones, ranging from $9.30 in South Carolina to $3.45 in Texas. The values for the other States stood : North Carolina, $7.41; Georgia, $5.93; Alabama, $1.31; Mississippi, $4.74; Louisiana, $3.92. The estimates of the cost of production were as follows: North Car- olina in 1880, three estimates, varying from 5£ to 10 cents, average 7 cents; in 1892, labor, provisions, and rent being cheap, the estimate was 3| cents on the best land, and 0.6 cents on land making 200 pounds of lint to the acre, which itself was a good deal above the average product of the region. South Carolina, eleven estimates, in 1880, rang- ing from 6 to lOf cents, placed the average cost at 8 cents per pound; in 1892, five estimates, running from 5£ to 7f cents, gives an average of 6.03 cents per pound. Georgia estimates for 1880 place the cost at 3 to 6 cents, under good management, with all necessary supplies. Alabama, 1880, estimates 8 cents ; 1892, cost 4| cents (exclusive of rent) and 7f cents. Mississippi, 1892, cost 0^ cents. These estimates, with one exception in Alabama, include charges for interest on plant, rent, and management, and would therefore indicate a fair profit. METAMORPHIC OR PIEDMONT REGION. The northern border of the pine-hills region touches a region extend- ing through North Carolina, South Carolina, Georgia, and Alabama, underlain by granite and kindred rocks, known as the metamorphic or Piedmont region. The southern limits of this region are marked by the falls of the rivers which, rising in the lofty Appalachian range that walls in its northern borders, pass through it on a steep incline and leave, it here for a quieter channel through the sands and alluvium of the lower country. The whole of this region is not planted in cotton. Two mountainous counties of northern Georgia produce none; in six coun- ties in western North Carolina 2,000 acres only were planted in it, while ten other counties there produce no cotton at all. The area of this region included in the cotton belt proper contains 32,000,000 acres, and produced in 1890 16.8 j)er cent of the entire crop. Only 10 per cent of this area was planted in cotton, being an increase of 2 per cent since 1880. Thirty-five per cent of the region is improved land and 80 per cent is in farms, being in both these regards a larger percentage than elsewhere in the cotton belt. Of the improved lands 30 per cent was in cotton in 1890, being 3 per cent more than in 1880. The acreage 238 THE COTTON PLANT. in cotton rose during the eleventh decade from something over 2,500,000 acres to nearly 3,500,000. The percentage of increase was greatest in South Carolina (35 per cent) and least in Alabama (20 per cent). The 400,000 additional acres in Georgia were only 21) per cent increase on what was in cotton there in 1880. The increase in North Carolina was 31 per cent, but it was accompanied by a decrease of 7 per cent of the crop. During the same period the crop of this region was increased 32 per cent. The increase of the crop, except in North Carolina, was gen- eral. Georgia showed an increase of 12 per cent, Alabama of 40, and South Carolina of 35. The number of acres required to make a bale remained very much the same. The average was 2.8 acres, Georgia alone showing greater production, making a bale to 2.7 acres, while it required 3 acres to do this in 1880. The per capita production of cotton also showed an increase, being 269 pounds of lint for each head of the X»opulation in 1890 against 228 pounds in 1880. The white population preponderates, only 42 per cent belonging to the colored race, a decrease of 1 per cent during the eleventh decade. The decrease was marked in South Carolina, Georgia, and Alabama, and was accompanied in each of these States by an increase in the per capita production of cotton. There was an increase of the colored population in North Carolina, and there a marked decrease in the per capita production of cotton is to be observed. In 1890 the 220,000 farms of the Piedmont region averaged 114 acres to the farm, a decrease in size from 131 acres in 1880. Only half of these farms were occupied by owners at the later date, while more than half (55 per cent) were so occupied in 1880. The percentage of small farms under 50 acres had also increased from 30 per cent to 40 per cent of the whole number. The larger porportion of renters paid a share of the crop, 38 per cent of the farms being rented on this condition and 12 per cent for a fixed money rent. In South Carolina and Georgia, where the colored populations were the largest, the greater percentages of rented farms and small farms were found, and the decrease in the average acreage of farms was greatest. The reverse of this was true of North Carolina and Alabama, where the proportion of the colored population was less. In the same connection it may be noted that mules far outnumbered the horses in South Carolina and Georgia and the horses outnumbered the mules in North Carolina and Alabama. The average of improved land to the work animal increased during the eleventh decade from 29 to 34 acres, and was general throughout this region. Three and seven-tenths bales of cotton were produced to the work animal in 1890, an increase of 0.7 of a bale over 1880. In South Carolina 5.5 bales to the work animal were made; in Georgia, 5.2; in Alabama, 4.3. Cotton culture in the South has not yet reached that stage of devel- opment where everyone has settled down on the best plan. Diversity of methods and of implements will be noticed on adjacent farms. CULTURE* OF COTTON. 239 Nevertheless, there is almost always some model system which is fol- lowed more or less closely and with greater or less success. In the Pied- mont region Mr. David Dickson, of Sparta. Ga., a very successful cotton grower and forcible writer on agriculture, set an example which has been very widely followed. He advocated deep breaking and subsoiling, saying that to stand a two weeks' drought a cotton plant must have 4 inches depth of soil and G inches depth of subsoil well broken, and for every additional week an inch more of soil, with the same subsoil- ing. lie did not recommend fall breaking, for the winters being usually mild and wet, plowed land was exposed to injury from washing and leaching, and was apt to run together closer than if it had not been broken. In cold, dry winters the reverse is true, but these are of rare occurrence in this latitude. The land should be broken as near the time for planting as practicable. Commence at the foot of the hill and cir- cle round on a level, finishing at the top. Eotation of crops was rest, cotton, corn, small grain, and rest again. Rows are laid off 4 feet apart with a shovel plow, running twice in .the furrow and leaving it 8 inches deep. Into this furrow drill the fertilizer and manure, running a scooter plow 5 inches wide on each side to cover it. Build up the bed by run- ning a turning plow, going 7 inches deep on each side of the scooter furrows and throwing the dirt over them. The bed is completed with a large 2-horse shovel running in the middle and bursting it out. Seed is drilled in with a planter. As soon as the plants are well up they are sided with a 22-inch sweep, running flat so as not to throw dirt, and then hoed, not chopped but scraped, the hoe never being raised more than 18 inches from the ground. The plants are left two or three together the width of the hoe apart, it being desirable to have eight stalks to the yard. Cotton thick m the row with good distance between the rows fruits earlier and better. The wing of the sweep should be turned up for the next plowing and a little dirt thrown to the cotton. The sweeps should be kept constantly sharpened, and should never run deeper than one-half to 1 inch. Two hoeings and ten furrows with the sweep, eight to side the cotton and two to split out the middle of the row, complete the working. This makes 1J days' work of a mule to the acre, going 10.G miles a day. In picking, hands were trained to pull out all the cotton from the boll at one movement of the hand. Pickers trained in this way gathered two to three bales a week for Mr. Dickson. He commenced using fertilizer in 1846, and was the first farmer in Georgia to do so, and his success induced many others to make enor- mous expenditures on this account. He did not think homemade manure any cheaper than commercial fertilizers, although he recom- mended its saving and use. It must be said that very few farmers get through their crops with two hoeings ; four are more frequently required. Crab grass, which pervades the whole cotton belt, is easily killed when young, but when it forms a stool and tillers not ouly is its growth rapid but its vitality is greatly augmented. " Crab " is a corruption of " crap," 240 THE COTTON PLANT. which is in turn the corrupted pronunciation of "crop" in some rural dialects. De Brahm, writing in 1752, says of it: New land produces scarcely auy grass, and one hoeing will do for the soason, but the grass conies and increases in such a manner that sometimes three hoeings are scarcely sufficient iu one season, and when this conies to be the case the planters relinquish these iields for pastures and clear new ground of its wood. It is on account of this grass, also, that most farmers prefer to hoe before plowing, to make sure of cleaning the drill of this fine grass, which is likely to escape notice if the dirt has been moved by the plow, so that until a shower has settled the dirt after plowing the drill can not be thoroughly cleaned by the hoe. The hoe also necessarily removes some dirt from the plant and should be followed by the plow to return it. Few farmers are able to plow only one-half inch, or even 1 inch deep, as Mr. Dickson did, although it would be very desirable to do so. If carefully measured, the depth of the cultivating furrow will be found to exceed 2 inches more frequently than it falls below it. The counties leading in cotton production are Mecklenburg, IS. C; Abbeville, S. C. (which stood third among the counties of the cotton belt in the annual production in 1890); Coweta, Ga., and Chambers, Ala. The value per acre of lands in farms increased during the eleventh decade from $0.07 per acre to $8.12. It stood as follows in the different States: North Carolina, $6.75 to $8.70; South Carolina, $5.91 to $8.75; Georgia, $6.01 to $8.38; Alabama, $4.17 to $5.43. The cost of producing a pound of cotton lint in the Piedmont region is stated as follows: North Carolina, 1880, four estimates, varying from 4.6 to 10 cents, average 6.2 cents; 1892, six estimates, varying from 5 to 8 cents, average 6.3 cents; South Carolina, eight estimates in 1880, varying from 5.71 to 8.25 cents, average 6.91 cents; 1893, eight esti- mates, from 5 to 8 cetns, average 5.7 cents; Georgia, 1893, six esti- mates, varying from 5^ to 8J cents, average 6.76 cents; Alabama, 1893, one estimate, 8 cents, which is the same as the cost given by Professor Smith for the whole State in 1880. SAND-HILLS REGION. The Piedmont region only touches the pine hills at certain points. Between them lies a belt of sand hills 500 to 800 feet in height, being often higher than the Piedmont counties immediately north of them. The sand-hills region extends through North Carolina, South Carolina, and Georgia. On reaching Alabama it passes between the Piedmont region and the black prairies, and, circling northwestward around the Coal Measures, spreads out in Tennessee and Kentucky. The total area is 6,000,000 acres, dG per cent of which is in farms and 5 per cent in cotton. It produces about one-tenth as much cotton as the Piedmont region, or 1.6 per cent of the total crop. During the eleventh decade the acreage in cotton increased 30 per cent and the crop 28 per cent. This was much more than the increase in the population, which was CULTURE OF COTTON. 241 only 17 per cent. Sixty-five per cent of the population is white, and here, as in other places where this occurs, if is found that horses are more used than mules; that a larger number of farms are occupied by owners; that the small farms are fewer, and that the size of farms (150 acres) is larger. Nevertheless, in most of these respects farming in the sand hills is changing in accordance with the general changes that are taking place in the cotton belt; farms are decreasing in size, small farms under 50 acres are increasing, owners occupy a smaller percent- age, and a greater number are being rented. The culture of cotton here resembles that of the adjacent regions, except that up to this date larger food and forage supplies and more stock have been raised here. THE PRAIRIE REGION. This region of the cotton belt includes the black prairies of Alabama, Mississippi, and Texas, the coast prairies of Louisiana and Texas, the gray-silt prairie of Arkansas, and the red-loam prairie of western Texas. The Texas prairies farther west, being sparsely settled, produce little cotton. The region thus indicated covers more than 65,000,000 acres, of which only 44 per cent in 1880 and 55 per cent in 1890 was in farms. The percentage of improved land rose from 12 to 27 per cent during the eleventh decade and the percentage in cotton from 4 to 6 per cent. In 1880 the prairies produced 16.3 per cent of the whole cot- ton crop; in 1890 they produced 20.6 per cent. The acreage in cotton had increased 48 per cent and the product 68 per cent, while the popu- lation had only increased 23 per cent, the per capita production of cot- ton rising from 277 pounds of lint to 381 pounds. The largest percentage of total area in farms was in the older States of Alabama (72 per cent) and Mississippi (74 per cent), and the smallest in the coast prairies of Texas west of the Brazos (24 per cent). The percentage of increase in the farm area was greatest in the red-loam prairies of western Texas (160 per cent), and there was a slight decrease in the prairies of north- eastern Mississippi, at the other extreme of this region. These red- loam prairies produced only 10,000 bales in 1880, and it was thought that cotton culture had at last reached its western limit here, but the returns of 1890 showed an increase to 74,000 bales, and the limits of cotton culture were moved still farther westward over a wide extent of country, which produced, however, only 6,000 bales in 1890. The most notable increase in the area in farms took place in the central black prairies of Texas (35 per cent), for it was accompanied by a doubling of the area in cotton and an increase of 146 per cent in the cotton pro- duced, the 249,000 bales made there in 1880 being increased to more than 600,000 bales in 1890. When it is remembered that only 7 per cent of the area of these black prairies in Texas is in cotton, against 18 per cent in the black prairies of the older eastern State of Alabama, an idea may be formed of the extent to which the production of cotton 1993— No. 33 16 242 THE COTTON PLANT. might be developed. There is good reason to believe that the 21,000,000 acres of the central black prairie of Texas could, if need be, produce the entire crop now grown in the South. Dallas County, Ala., stands first among the cotton-producing counties in the prairies of that State, with a crop of 42,000 bales in 1890; but this was far below the crop produced as far back as 1800, which was over 63,000 bales. So, too, Hinds County, Miss., led the prairie coun- ties there, with a crop of 42,000 bales m 1800, which again fell short of the crop of 54,000 bales which it produced in 18G0. Ellis County, Tex., which made only 2,000 bales in 18G0, led the black prairie counties of that State in 1890 with a crop of 42,000 bales; in 1892 it made a crop of 50,000 bales, and in 1894 more than 59,000 bales were shipped from Waxahachie, the county seat, and the best estimates place the crop of the county for that year at 100,000 bales. This gives it the distinction of leading all the other counties of the cotton belt in the amount of cot- ton produced. In 1880 40 per cent of the population of the prairies was colored ; in 1890 the percentage had fallen to 39. It stood, in Alabama, 78 per cent; in Mississippi, 51 per cent; in Louisiana, 44 per cent; in Arkansas, 37 per cent; in Texas, 18 per cent. The following table shows the figures given in the Eleventh United States Census in regard to land tenures in the prairie region, and also the size of farms, motive power in agriculture, value of farms per acre, and the acres required to make a bale : Statistics of cotton culture in the prairie region of the cotton belt. Locality. I. Black prairie, Texaa Gray-silt prairie, Arkansas... Red-loam prairie, Texas Coast prairies, Texas Coast prairies, Louisiana II. Black prairie, Alabama Black prairie, Mississippi Northeast prairie, Mississippi 173 118 401 220 120 98 119 102 Or- 03 s Number of counties in which white or colored and horses or mules, outnumbered. $12.19 11.83 5.33 9.82 11.83 6.40 5.37 7.79 2.4 2.3 2.3 2.4 2.6 3.0 2.7 4.4 These facts point in one direction, and the general conclusions are that where the owners of farms largely occupy them (1) the white population generally predominate ; (2) horse power is used in preference to mule power; (3) farms have a larger acreage; (4) farms under 50 acres — that is, small farms — are fewer; (5) fewer farms are rented; CULTURE OF COTTON. 243 (6) the value per acre of land is greater; (7) the production of cotton per acre is greater. Where fewer owners occupy their farms, the facts are the reverse of these several statements. The lower value of land in the red-loam prairie of western Texas apparently forms an exception to these conclusions. When it is pointed out that this is a sparsely peopled country;, newly settled, it ceases to be an exception. The farm acreage, too, on the gray-silt prairies is 1 acre less than the highest acreage in the second series. The explana- tion of this, if any is needed, is that these light soils have been less attractive to capital than the rich alluvium to be found in nearly every county of well-watered Arkansas ; that they tally so well with the more fertile sections tends to confirm the general conclusions. In making this collocation of facts it is not meant to emphasize any one as the only cause of all the others. They go together, and it may be safely said that the difference in cotton production and in the value of farms does not result from any natural conditions, such as greater or less fer- tility of the soil or a more or less favorable climate. The facts already given as to the cotton production of the leading cotton counties of Alabama and Mississippi in 1860, when agriculture was less advanced, show that in natural fertility they were not inferior to the black prairies of Texas, now the most productive in the prairie region.1 If they have become so since it must result from the methods pursued by the men who cultivate them, either from choice or the force of circumstances. The changes which occurred in the rural economy of the prairies during the eleventh decade were similar to those taking place through- out the cotton belt. The number of farms occupied by their owners decreased; the number of small farms increased and the average size of farms diminished; the amount of improved land to the work animal increased, and the percentage of the colored population was less. The increase in the acreage of cotton was accompanied by a very much greater increase in the crop, which was a marked exception to the general rule. The number and variety of implements recently introduced in cotton culture here, especially in the prairies of Texas, is very much greater than elsewhere in the cotton belt. The planting year commences there December 1 — a month or more earlier than in the East — and a good deal of work is done before January 1. If all the cotton has been picked, the first operation is to dispose of the cotton stalks. This is usually done with a stalk chopper. A number of patented implements are used for this purpose. They consist of five to seven steel knives over 2 feet in length, bolted to iron arms which revolve in a frame swung from auother frame supported on two wheels. Two horses pull this machine along the cotton row, and the stalks are chopped off by the revolving knives 2 to 1 inches above the ground and cut into pieces. 'Tenth Census of the United States, 1880, Vols. V and VI. Cotton Production, Parts I and II. (See also chapter on climatology and soils.; 244 THE COTTON PLANT. A driver rides, and adds his weight to the blow of the knives. Some- times a double stalk chopper is used, which takes two rows at a time, and is pulled by three horses, thus saving one hand and a horse. The work is very thorough. Cornstalks are removed by running a heavy roller along the rows, breaking them down, when they are raked by a very large wooden rake, pulled by horses, at right angles to the direc- tion in which the roller moved, piling them in windrows, to be burned. With both these implements a seat is arranged for the driver, and he rides. Breaking, and, in fact, almost all the work of tillage and culti- vation, is done with two horses. Several varieties of steel plows with steel beams are used. A large disk plow is being introduced. It con- sists of one to four heavy steel disks revolving in a heavy frame. Four to eight horses are required to pull it, and it breaks from 2 to 7 acres of the tough prairie land to the depth of 8 inches in a day. Often, however, the preparation for planting cotton consists only in throwing two 2-horse turning-plow furrows into the old alley and completing the preparation by running a large 2-horse double-mold plow in the last year's bed, making in all only three furrows to the row. It is thought best to do this work early, so as to allow time for the beds to settle, which are sometimes freshened up by running a cut- away harrow across them just before planting. A number of excellent cotton-seed planters are used, and they are implements of much greater precision than the Dow-Law planter and its modifications, used in the East, and their cost is proportionally greater. The object aimed at is to deposit the seed regularly in a single line, with barely a space between each, which, while it saves seed, obviates their coming up in bunches, difficult to thin, and lifting the earth and thus exposing the teuder stems to the vicissitudes of the spring weather. Practically no fertilizers are used, and very little of the manure from the many thou- sands of beeves fattened here yearly on cotton-seed meal and hulls is utilized. Some farmers say that manures cause the crop to fire in dry seasons, and that the natural fertility of the soil suffices, lands that have been planted successively for forty years producing better than when fresh. There is no doubt that these stiff soils are pulverized and rendered lighter and more productive by some years of tillage, but the old farmers admit that the effect of continued cropping is beginning to show in a somewhat diminished yield in places. Biding 2-horse cul- tivators have been used for many years. Before their use cotton rows were made 4 feet apart. Since their introduction it has been found that these cultivators clean the rows best when they are spaced 40 inches, and that is now the most common width. Various shares, to suit the circumstances of the case, are used with these cultivators. The first workings, which are deepest, are given with 6-inch sweeps, afterwards 12-inch sweeps are used, and later sometimes even larger ones. The culture is tolerably flat. The double shovel is sometimes used, and it is almost the only 1-horse implement used on most of the CULTURE OF COTTON. 245 black Texas lands. Some farmers say they would not know how to hitch up a 1-horse team, so little are they used. It is calculated that these sulky cultivators should plow 10 acres a day, but in practice they rarely do more than 7. Two to three hand hoeings are given, and plowing, under the best culture, is done six to seven times in the sea- son, more to stir the surface than to kill grass and weeds. Grass is not so troublesome here in favorable seasons (moderately dry) as else- where. Coco, or nut grass, is not much feared, but there is great apprehension regarding Johnson grass. On some farms where it- appears it is fenced off to prevent its spreading and treated with heavy doses of arsenic, which kills the land as well as the grass. This would seem hardly necessary, as this grass is readily subdued in the East by close pasturing with stock, and the prairies of Texas themselves fur- nish an object lesson of the eradication of the native perennial grasses by continuous grazing. All the lands in Texas are fenced in with wire, woodland as well as that under cultivation, and cattle and horses everywhere feed on the stubble after the crops have been gathered. Picking cotton is a heavy item of expense, the usual price being 50 cents per hundredweight of seed cotton. Cotton pickers here are more expert than elsewhere; children 6 years old sometimes pick 100 pounds, and girls of 9 years have picked as much as 200 pounds. First-class pickers average 500 to 600 pounds a day, and as much as 800 pounds occasionally. A white hand was timed in 1894 and picked 60 pounds in an hour, or a pound a minute. Such results require a very abun- dant blow of cotton, and the picking is not very neat, a good deal of trash, such as the hulls of bolls and stems, being mixed with the cot- ton. A variety of " storm proof" cotton so called because it is not easily blown out by winds, is preferred. Nevertheless, in 1895, it was not unusual to see as much as 200 to 300 pounds to the acre of the crop of the previous year fallen out on the ground. A good deal of the crop, also, was left unpicked, not only where the crop was heavy, but almost in the same proportion on the poorer lands. The explanation given was that the tenant farmers did not think it would pay, at the prevail- ing low prices, to pick, gin, and pack what was in the field for their share — three-fourths of it — and so preferred to send their children to school, with which no doubt the children heartily agreed, for picking cotton in cool weather is not attractive work, being an exercise too ligh to keep the blood warm. Cotton is not infrequently left in piles in the field after being picked, a practice long abandoned elsewhere, but less ruinous here on account of the dryness of the climate. It is more usual to see a white, canvas-covered wagon left in the held, its pole elevated, to which steelyards are suspended for weighing the cotton, which is after- wards loaded into the wagon and left there until it is hauled to the gin. It is not feared that the cotton thus exposed will be stolen, which is quite different from the case in other States, where it has been found necessary to pass stringent laws punishing the pilfering of seed cottou. 246 THE COTTON PLANT. There is very little storage for cotton in Texas. When it is not carried immediately from the gin to the compress and loaded at once on plat- form cars for transport, the bales are placed on edge in open fields, protected only by a wire fence, and exposed to rain, wind, and dust. The wastefulness of cotton culture is great in Texas, owing to the great abundance of the staple. There seems to be a waste in horse- power, such small areas being tilled to the horse, and also in wages paid to labor, which have ranged from $10 to $18 a month, with shelter, fuel, and rations; fully double what they are in the East. It was common to give $10 a month for ten months and $18 a month for the two months during the picking season. Owing to the low prices a reduction is being made to $10 for ten months and $20 for the months at picking. The fact is, very little labor is hired in Texas. By far the larger portion of the land is rented out for money rent or worked on shares, the latter being done to a much greater extent than is exhibited by the census returns. The estimates of the cost of production in 1880 were as follows: Mis- sissippi, 1 estimate, 11 cents per pound of lint; Arkansas, 10 estimates, varying from 3 to 8J cents, average 6.2 cents; Texas, black prairie, 14 estimates, 34 to 9£ cents, average 5.3 cents, coast prairies, 7 estimates, from 44 to 94 cents, average 6£ cents, red loam prairie, 2 estimates, each 44 cents. Notice is again directed to the fact that where a country plants only a limited area in cotton, although that country is appar- ently less favorable for this crop than other more fertile localities, nevertheless the estimates of cost from the limited area are almost always the lowest — an indication that were this culture restricted everywhere to narrower limits cotton might be grown everywhere at a much lower cost. In 1891 circulars were sent from the agricultural experiment station in Texas requesting the leading farmers in various parts of the State to keep accounts of the actual cost of cotton production in 1892 and furnish the figures to the station. All replies not supported by actual results were discarded. The replies from seven careful, practical cotton growers are summarized in the table below, Nos. 1 to 7. The officers of the station undertook themselves to obtain the same infor- mation on the State's farms, and these results are summarized in Nos. 8 to 12. In the statements marked / fertilizers were used; in the others the results were obtained from the natural fertility of the soil. In only one instance (No. 2) was any charge for management entered. No. 6 is reported by Jeff Wellborn, who gathered the crop of l,r>00 pounds of seed cotton from the acre with a cotton picker at a cost of 10 cents per hundredweight. Figures marked with an asterisk (*) indicate loss. Where the price of the seed is not credited it is because they were exchanged to pay for the ginning and packing. The price of seed in 1892 was $0.50 per ton and in 1894 it was $6, but this difference is not charged in the tables. The first section of the table shows the actual CULTURE OF COTTON. 247 cost of the crop, value of product, and profit for 1892. The second sec- tion gives the data of 1892 estimated at the average price (4£ cents) the crop would have brought in 1894. Cost of cotton production. .Number. 1/ 2 3 4 5 e 7 8/ 9/ 10/ U/ 12 Average 1892. Cost of T,int Price cultivat- ! ,„u Value per 1 , cotton ,. j ' ■• mslacre , ot seed, pound in rnttv.n PrOUUteU. f „ * in cotton. $30. 55 22.32 13.68 14.71 21.31 13. 85 14.15 34.90 29. 36 27.17 31. 23 18.84 22. 02 Pounds. 650 418 | 250 i 250 ; 426 ; 500 250 I 544 i 484 423 502 283 $4.00 3.00 Profit per acre. 3.25 3.50 3.12 2.71 3.22 1.82 Cents. $31. 95 16.18 . 7.57 4.39 10.64 24.40 7.10 20.96 20.34 16.25 20.00 9.21 1804. 415 15.77 $31. 63 20.77 10.62 12.22 18.00 24.50 10.62 26.62 23.69 20.69 24. 56 13.85 Profit and loss. $1. 08 *1. 55 *3. 06 *2.49 *3. 30 10.65 *3.53 *8.28 *5.67 *6.48 *6. 67 *4.99 *2. 85 Rent. $4.00 3.00 3.50 3.96 7.98 4.00 3.50 3.00 3.00 3.00 3.00 3.00 Proceeds per acre, rent deducted. $5.08 1.45 .44 1.47 4.68 14.65 *.03 *5.23 *2.67 *3.48 *3.67 *1.99 3.73 Excluding No. 6 on account of the very exceptional method by which the cost of picking was reduced, the average cost of production varies from 1.6 to 6.1 cents. These were good crops, making the unusual average of 115 pounds of lint per acre. At the prices of 1892 they showed an average profit of $15.77 per acre. The same crops made at the same cost would have shown an average loss per acre at the prices (4J cents) of 1891 of $2.85. The rents were 13.73 per acre in 1892, and if they are omitted from the charges against the crop in 1891 the cotton growers would have had a surplus above actual expenses of 88 cents an acre. No charge for man- agement was made in any statement except No. 7. The profit in 1892 was large — larger perhaps, thau any profit from any staple cultivated on so extensive a scale. Adding in the rent, there is a clear surplus of $19.50 an acre. Counting the land, buildings, fences, stock, implements, and working capital at 8195 per acre, it was paying 10 per cent (the legal rate of interest in Texas) upon them. Taking the average of the black prairies of Texas, the investment could not have exceeded $31; double that, on the assumption that these were choice lands, and still the profit was over 30 per cent. The commissioner of agriculture of Texas in his report for 1892 states that with a population of 2,500,000 and 250,811 farms there were only 57,012 farm laborers working for wages, which explains the high wages paid for farm labor, the average being 813.S9 per mouth. If each of these farms averaged three cotton pickers and they should only pick 100 pounds each a day, it would amount to 50,000 bales daily, or in the 100 days of a short picking season 5,000,000 bales could be gath- ered if the labor were effectively distributed. 248 THE COTTON PLANT. OAK AND HICKORY REGION. This region lies on the eastern border of the black prairies of Texas. It extends over eastern Texas and northwestern Louisiana, and south- eastern Arkansas to the Mississippi swamps. Its counterpart in the east reaches westward from the northern portion of the black prairies of Alabama and Mississippi to the table-lands and bluff region that rise from the swamps of the Mississippi River. Four counties in west- ern Alabama and three in northeastern Mississippi, just east of the black prairies, are included in this region. It covers nearly 43,000,000 acres, about half of it lying in eastern Texas, and in 1890 it produced 14.4 per cent of the entire crop, and in 1880 it produced a fraction of a per cent less. In 1890 52 per cent of this area was included in farms (over 129,000 in number), and 21 per cent either had been or was under cultivation, nearly 4,000,000 acres of fresh land being brought in during the eleventh decade. This increase was naturally greater in the newer Western States than in the older Eastern ones, rising from 31 per cent in Alabama to 129 per cent in Louisiana. This region reported in 1880. a large amount of land turned out of cultivation, which is included as improved land in the returns of 1890. Such abandoned fields were reported as 20 to 90 per cent of the lands once under cultivation in Mississippi, as 15 to 30 in Arkansas, as 10 to 50 in Louisiana, and as 1 to 50 in Texas. The growth of oak, hickory, and short-leaf pine furnish- ing material for an extensive lumber business, is evidence of the orig- inal good quality of these soils, which immigration has passed over in its movement westward. The area planted in cotton was 7 per cent of the entire area and 32 per cent of the improved land, and increased 45 per cent during the eleventh decade, while the cottou crop only increased 30 per cent. The percentage of increase of acreage in cotton and the percentage of increase of the crop stood as follows in the different States : Alabama, increase of acreage 29 per cent, increase of crop 27 per cent; Mississippi, increase of acreage 14 per cent, increase of crop 12 per cent; Louisiana, increase of acreage 50 per cent, increase of crop 32 per cent; Texas, increase of acreage 52 per cent, increase of crop 44 per cent; Arkansas, increase of acreage 50 per cent, increase of crop 15 per cent — everywhere the same story, so often repeated, that an ncrease of acreage is not accompanied by a proportional increase of crop, but inevitably by such an increase as reduces the price of the product in a ratio quite as large as that of the increased cost in culti- vating the larger area. Within the last thirty years marked changes have occurred in the localities producing the largest amounts of cotton. In Texas in 1860 San Augustine County produced 31,000 bales, which was not only more than any other county in the oak and hickory region, but more, also, than any other county in the State; but in 1890 this county produced only 4,000 bales. The attraction of the western prairies, and the fact that the county lies off the lines of railroad CULTURE OF COTTON. 249 communication, in part account for this, but this being a well- wooded sec- tion lumbering has temporarily diverted attention from agriculture. Bossier Parish, in Louisiana, led the parishes of the region in I860 with 40,000 bales, but produced only 29,000 in 1890. Pickens County, Ala., led in that State with 29,000 bales in 1800, which were reduced to 18,000 in 1890. Kemper was the leading county of the region in Missis- sippi, with a crop of 15,000 bales in 1800, which fell to 11,000 in 1890. Arkansas alone is an exception; Jefferson County made 28,000 bales in 1860, and increased the crop to 47,000 in 1S90. During the eleventh decade the product per acre decreased in this region, and it recpiired 2.8 acres at the close of the period to make a bale which had been done on 2.5 acres at the beginning. This decrease in the yield per acre was general. There was also a small decrease in the product of lint per capita of the population, which stood in 1880 at 358 pounds and in 1890 at 351 pounds. There was an increase in the population of 38 per cent, which, while it was less than the increase in the cotton acreage, was greater than the increase in the cotton crop. Fifty-one per cent of the population was colored in 1890, which was 1 per cent less than in 1880. The high, est percentage (5o) of colored population was in Louisiana, which showed an increase of 2 per cent during the decade, and the lowest percentage (30) was in Alabama, showing a decrease of 10 per cent since 1880. There was a slight percentage of increase in Arkansas, and a slight decrease in Mississippi and Texas. The average size of the farms decreased during the eleventh decade from 170 acres to 136 acres. This decrease was general, being greatest in Louisiana, where the size of farms fell from 206 acres to 124 acres, and in Arkansas, where the decrease was from 167 to 141. These two States had an increase in the percentage of colored population. In the other States, which showed a decrease in the percentage of the colored population, the decrease was considerably less. The number of small farms of 50 acres and under increased from 31 to 34 per cent, and the number of farms occupied by owners decreased in a somewhat greater proportion from 67 to 62 per cent. The decrease iu the average size of farms, the multiplication of very small farms, and the decrease in the occupancy of farms by their owners, stand parallel to each other, and parallel also to a large extent, but not wholly, with the increase or decrease of the percentage of the colored population in the several localities. The significance of the facts are that the white proprietors are turning over the management of the lands to the small colored farmers, and that "40 acres and a mule" has ceased to be an idle dream of the freedman. Horses on the farms in this region exceed the mules, the latter out- numbering them in only 8 counties in Louisiana, Mississippi, Arkansas, and Alabama, out of the 89 counties comprising this region. The average number of acres of improved land to the work animal was 23 250 THE COTTON PLANT. in 1890, against 16 in 1880; the largest number was 36 acres, in Louisi- ana, and the smallest 18 acres, in Texas and Louisiana. An average of 2.4 bales was produced to the work animal, ranging from 3.7 bales in Louisiana to 0.9 of a bale in Texas. There is no regular system of rotation of crops in Texas. Cotton is alternated with corn on fresh land for nine or ten years and then the land is planted continuously in cotton until it ceases to produce 600 or 700 pounds of seed cotton to the acre, when it is abandoned and fresh land planted. On the Arkansas oak lands cotton is not infrequently planted for ten years in succession. Where rotation is practiced cot- ton is followed by corn and then oats, when peas may be planted or cotton follow on the oat stubble. There is seldom any fall plowing, and the usual practice is, after knocking down the cotton stalks or pulling them up and burning them, to lay off in the middle of the old row, if manure or fertilizer is to be used; if no manure is applied the land is simply bedded up on the old middle. The land being easy of culture for the most part, this is done with a 1-horse plow. Cotton- seed planters have come into very general use since 1880. When the plant appears the rows are barred off with a scrape or turn plow, the cotton chopped out, and the dirt thrown back to it either with a turn or shovel plow and sometimes with a scooter. The subsequent plow- ings are done, by preference, with sweeps, but in grassy seasons turn plows and shovels are used. Two to three hand hoeings and three to four plo wings are required before the crop is laid by, early in August. Deep cultivation is said in Mississippi to prevent the plant from run- ning to weed and to favor fruiting. Close and deep plowing is prac- ticed in Louisiana to effect this. In Texas, root pruning is thought efficacious; it is done by running a deep scooter furrow on one side of the row and after an interval treating the other side in the same man- ner. Topping, early planting, and rapid cultivation with the plow are recommended everywhere to check growth and cause the plant to fruit. Late cotton is thought more apt to run to weed, and also cotton that is planted close in the drill. It will be seen that all these practices are directly the reverse of what is done in Georgia. In Arkansas, also, it is thought boiling is promoted by planting thick in the drill, and run- ning to weed prevented by shallow cultivation. It would be difficult to reconcile such conflicting methods effecting the same object. What experience has taught, however, is not to be lightly set aside, and there might be risk in denying that root cutting in the lower latitude of Texas did not modify the perennial character of the cotton plant, dwarf its growth, and cause its energies to be thrown more toward bearing fruit; or that this perennial character being already restrained by the climate of the higher latitude of Arkansas, fruiting would fol- low on a more conservative culture. The value per acre in farms of these lands rose from $4.22 in 1880 to to $5.94 in 1890. This value was greatest in Texas ($7.07) and least in CULTURE OP COTTON. 251 Alabama ($3.03). The average value of all the agricultural products per acre fell from $10.09 in 1880 to $7.86 in 1890. The estimates of the cost of producing a pound of lint cotton in 1880 were : For Arkansas, 9 estimates, ranging from 1 to 9 cents, average 6.22 cents; for Louisiana, 7 estimates, ranging from 4i to 8 cents, average 6.8 cents; for Mississippi, 1 estimate, 4J cents; for Texas, 12 estimates, ranging from 3h to 9£ cents, average 6.1 cents. The estimates of the cost of producing a pound of lint cotton in the oak and hickory lands in 1893 are: For Louisiana, 6 estimates, varying from 5 to 10i cents, average 6.4 cents; for Arkansas, 5 estimates, ranging from 5£ to 9 cents, average 7 cents. BLUFF AND BROWN LOA^I TABLE-LANDS. This region lies east of the flood plain of the Mississippi Eiver and extends eastward to the oak uplands. It reaches from Paducah, on the Ohio Eiver, to Baton Eouge, in Louisiana. This region covers more than 11,000,000 acres and produced in 1890 6.8 per cent of the entire cotton crop, having produced in 1S80 10.4 per cent. During this period the cotton acreage increased 21 per cent, very nearly double the rate at which the population had increased, and the crop decreased 13 per cent. The area in farms increased from 74 to 77 per cent, it being the most thickly settled region of the cotton belt, except only the Piedmont. The improved lands increased from 30 to 36 per cent, and the lands in cotton from 10 to 12 per cent of the whole area. The percentage of improved land in cotton (35) was the same as in 1880, but it required 2.9 acres to make a bale, which was made on 2.1 acres in 1880. The largest yield was in Louisiana, where 2.1 acres produced a bale in 1880 and 1.6 in 1890. The cane hills of Mississippi made nearly as much, producing a bale to 1.7 acres in each year. The summit in Tennessee made the lowest yield, 2.3 acres in 1880 and 4.2 acres in 1890 being required to make a bale. The production of cotton per capita was 431 pounds of lint in 1880 and 347 in 1890. Fifty-two per cent of the population was colored in 1890, the per- centage having decreased by 2 since 1880. There was a decrease in the proportion of colored to white in Tennessee and Louisiana and an increase in Mississippi, where it formed 69 per cent of the population. Horses outnumbered the mules on farms in 38 out of the 76 counties in this region. The general average was 19 acres of improved land to the work animal, ranging from 23 acres on the table-laud of Mississippi to 17 on the bluff of Louisiana. Two and three-tenths bales were pro- duced to the work animal in 1890 against 3.2 in 1880. The cane hills in Mississippi produced the most, 5.3 bales to the work animal, and the summit in Tennessee the least, 0.7 bale. The farms averaged 114 acres in 1880 and fell to 106 acres in 1890. The proportion of small farms remained the same, but the percentage occupied by owners fell from 49 to 46. More were rented for a share of the crop than for a fixed money rental. 252 THE COTTON PLANT. Broadcast breaking of the soil is rarely done, except on stubble land or land that has been out of cultivation for some time. After cleaning off the stalks, the fertilizer or stable manure, when any is applied, is put in a shovel furrow in the old alley and bedded on. Cotton is planted in ridges 3£ to 4 feet, according to the strength of the land. Cultivation commences as soon as the plant is fairly up by barring off the bed with a scrape or turn plow and sometimes with a bull tongue, The cotton is then hoed out and the dirt returned to it with a turn plow, shovel, or sweep. Sweeps are used for the later plo wings, and the cotton is usually hand hoed before each plowing. The weed grows larger here than in Georgia and the Carolinas and is not so heavily fruited, the seed are large, and the ratio of lint is less. It is said, also, that these seed yield less oil. Under these circumstances the farmers are especially desirous of checking the growth of the plant and induc- ing it to fruit, particularly in wet seasons. Some admit that there is no remedy for this, but many are persuaded that their methods are successful in attaining this end. One recommends shallow culture, cultivating as much as possible with the hoe, and early laying by. His neighbor is equally sure that deep culture and cutting the roots stops the overgrowth and promotes fruiting. Topping is generally spoken of as good, and stable manure and superphosphate are recom- mended. One set advise that the dirt be thrown to the plant, and another that it be drawn away. Early planting and early thinning are thought by others to be effective. The value per acre of land in farms in this region varied in 1890 from $5.64 in Louisiana to $10.92 in brown loam of Tennessee. The average for the region was $9.12 against $8.16 in 1880. In 1880 four estimates of the cost of production in Mississippi varied from 5i to 10 cents and averaged 7.2 cents per pound of lint. The estimates in Tennessee varied from 3£ to 10 cents and averaged 8£ cents. The estimates of the cost of production in 1893 were, for Louisiana, 6.23 to 7 cents per pound of lint; for Mississippi, from 7 to 10 cents (average, 8.4 cents) ; for Tennessee, 7 cents. THE ALLUVIAL REGION. Extensive areas of alluvial lands occur in every Southern State and in almost every region of every State. For example, in South Carolina there are 400,000 acres of such lands below the falls of the rivers, all of it susceptible of drainage, with much that might be irrigated, while scarcely any of it is now cultivated. In such well-watered States as Arkansas and Louisiaua there is scarcely a county in which corre- spondents do not mention the cultivation of bottom lands; but when the alluvial region of the cotton belt is spoken of it is meant to desig- nate the bottom lands of the Mississippi, the Red, and the Brazos rivers. The region thus located contains very nearly 30,000,000 acres, CULTUKE OF COTTON. 253 only 34 per cent of which is in farms, 12 per cent improved, and 6 per cent in cotton, producing 14.4 per cent of the entire cotton crop in 1890 and 14.6 per cent in 1880. Thirty-five per cent of the improved land is in cotton, and no change in this regard is shown since 1880. The aggregate increase in cotton acreage is large, 01 per cent being added to the acreage of 1880, which was accompanied by an increase of yield of only 29 per cent, something more, however, than the increase of the population, which was 25 per cent, raising the per capita pro- duction from 368 pounds of lint in 1880 to 393 pounds in 1890. The average of the whole region was a bale to 1.8 acres, while in 1880 this amount was produced upon 1.4 acres, a decline, but still a yield per acre far ahead of any other region of the cotton belt. The highest average yield in 1890 was in the Louisiana bottoms — a bale to 1.4 acres; and the highest of recent years was on these same bottoms in 1880 — a bale to 1.2 acres. The lowest average yield in 1890 was on the alluvium of the Red River — a bale to 2.9 acres — while the same lands produced a bale to 1.8 acres in 1880. Fifty-one per cent of the population is colored, a decrease here, as in every other region except the oak and hickory, since 1880. It stood then at 57 per cent. The percentage of colored population stood high- est in the State of Mississippi, where it had increased from 79 per cent in 1880 to 83 per cent in 1890. It stood lowest in the Brazos and Eed river bottoms, declining from 31 per cent in 1880 to 29 per cent in 1890. This should be noted, for some of these lands have the reputa- tion of being the least suitable of all for a white population. Their healthfulness has been very greatly improved on the Brazos by sink- ing artesian wells. In some places pure water can be obtained at a depth of 300 feet and at a cost of $75 to the well. In the swamps of Louisiana, Mississippi, and Arkansas the first settlements were made by wealthy planters with numerous negro slaves, and here the colored populations, remaining where they were planted, on the most fertile soils, preponderate to this day. The later settlements on the alluvium, about Crowleys Ridge in Arkansas, along the Red River in Texas, and in the tide-water alluvium of Louisiana, were made chiefly by whites, and there they outnumber the colored race. As the whites preponderate in the extreme south of this region as well as in the north, and sometimes in the central portions, as in Avoyelles Parish, La., the distribution of the races can not be attributed to climatic causes. There is no foundation for the prevalent belief that heat or moisture or malaria has caused the segregation of the colored population. It was effected solely by economic influences, and has been maintained by a race prejudice which has deterred white immi- grants from occupying many of the most fertile sections of the South. Mules outnumber the horses on farms in the alluvial region, and this is especially true where there are large colored populations as in Mississippi and Louisiana; horses, on the other hand, outnumber the 254 THE COTTON PLANT. mules on the alluvium of the St. Francis bottoms in Arkansas and in Texas, where the proportion of colored to white is much less. There were 1(5 acres of improved land to the work animal in 1890 against 13 in 1880. This increase was general in every section. The largest acreage was in the State of Mississippi (19 acres against 17 in 1880) and the least in the Eed River section in Texas (14 acres against 11 in 1880). There was an average of 3.G bales made to the work animal in each year. In Mississippi and in the alluvial region north of lied River in Louisiana 7.4 bales were made to the work animal, and in the tide- water alluvial region of Louisiana only 1 bale. The average size of farms decreased from 171 acres in 1880 to 117 acres in 1890. The percentage of farms of 50 acres and less was very considerably increased. At the same time the percentage of farms (as is invariably the case) occupied by owners showed a decrease from 49 to 4G per cent. The increase in the percentage of rented farms was especially noticeable where the colored population preponderated, rising from 48 per cent in 1880 to 72 per cent in Mississippi, but only from 44 to 50 per cent in the Red River section of Texas. There is reason to doubt whether the figures of the Eleventh Census represent the changes of this character existing in 1894. A much larger propor- tion of the lands must be rented, chiefly for a share of the crop. So little land is worked by hired labor that in some sections managers of large estates have no rate of wages, tenants to the number of several hundred cultivating the land under various modifications of the share system. Few agricultural regions anywhere present so fine an appearance as these alluvial lands. The traveler may pass for miles through broad, level, neatly cultivated fields, extending on either hand from 1 to 2 miles and walled in by a heavy growth of tall and stately forests. The cotton planters are gone, but the cotton plantations remain with their comfortable quarters, formerly for laborers, now for tenants, so numer- ous as to have the appearance of a widely scattered, but continuous village. The farmstead in a group of pecan trees, with the manager's neat cottage, the large stables, and the ginhouse fitted with the most modern improvements, have a prosperous look. At rare intervals the handsome residence of the old owner may be seen, but it wears a deserted look, and it is very seldom that either he or his family occupy it, even for a short time. They have moved to the towns, and he has taken up a profession, or become a banker, merchant, cotton factor, or engaged in some other occupation which helps him to support his plan- tation. There is everywhere a large amount of fresh land cleared. The tenants' rents are remitted for such clearings, which cost the owner little outlay. The broom sedge of the abandoned old fields, so conspic- uous a feature in the landscape of the uplands, is no longer seen here. Outside of these clearings, the chief improvements are those which have been made in the levee system of the Mississippi River. Erevious to CULTURE OF COTTON. 255 the year 1859 their construction was chiefly in the hands of local boards of planters, and no decided results were obtained. In 1S79 the Missis- sippi Eiver Commission was organized, and the Congress of the United States appropriated $4,000,000 for the work. Since that date great progress has been made in the efforts to control the flood waters of the Mississippi, the work being carried on by Congress in conjunction ■ with State and local boards. The latter impose local taxes, usually so much a bale for each bale of cotton raised in the district, and Congress adds a sum equal to the tax raised. Whether this work when com- pleted will fulfill all the expectations that are entertained for it remains to be seen. These lands have been considered inexhaustible in their fertility, but it must be remembered that this fertility has been from time to time renewed by the floods that are being walled out. In other alluvial regions farmers who have dammed out the river freshets have found, after some seasons of cropping, that their lands were becoming less fertile. There is no regular system of rotation of crops. Land that has been planted in cotton for forty successive years in Bolivar County, Miss., produces just as well the last year as it did the first. Nevertheless, a desultory rotation is sometimes practiced. Cotton is followed by corn, and after corn oats are put in, a crop of peas being taken after the oats, with cotton again the next year. In Louisiana, sugar cane and potatoes are sometimes added to these other crops, but it is seldom that cotton is left out for more than one year. The implements of agricul- ture used are like everything else here — of the latest and most improved pattern. They are similar to the implements noticed when speaking of the Texas prairies. A new sulky cultivator with disks — three on each side of the row, which it straddles and works at one operation — is being introduced in the place of plows, and much is expected of it. As a rule, the early cultivation is deep. The turn plow or scrape is used to throw the dirt from the cotton, and after it is hoed and thinned to return the dirt to it. There are three to four hand hoeings, and it is plowed as frequently, sweeps and various harrows being used later in the season. Besides the universal crab grass, coco, or nut grass (Gyperus rotundus), gives great trouble. After being shaved off with the hoe or plow it resumes its growth immediately, and adds considerably to the cost of culture. Harrowing is thought the most effective means of tearing the roots and nuts, and the effort is made to cover it with dirt from the plow instead of cutting it off, for, in the first case, it is found that leaves rot off before it grows again. The land is generally left clean and nearly level when the crop is laid by. The efforts to promote fruiting and prevent the plant from running to weed are the same as those already noticed — as various and as conflicting. A single subsoil colter furrow run in the middle of the row is the only new sugges- tion, both sides of the row having been treated in a similar way by others. A few admit that there is no remedy when a warm, wet season supervenes. 256 THE COTTON PLANT. The usual wages are 50 cents a day, or 50 cents per hundred weight, for picking cotton. Very few laborers are hired by the month or the year. As has been said, most of the land is rented for money — $3 to $5 an acre, formerly for as much as $10 — or is worked on shares of the crop. The laborers are furnished with a house, fuel, and garden, and when the law requires stock to be fenced in a pasture is inclosed for their stock. There is no charge for this, and only such supplies as are advanced are charged. In some sections it is customary to pay $5 an acre for the area of cotton a hand plants and cultivates up to the time of laying by. Picking is paid for extra, and so is any other work out- side of the acres contracted for. The very low prices of the last season have induced employers to discuss the propriety of reducing this wage to $3.50 an acre. When it is remembered that field work does not commence often until March, that all other work is paid for at the rates of day labor,, that unlimited hunting and fishing is allowed, in an excel- lent game country, and that all the perquisites mentioned are enjoyed by the laborer for an entire year, it does not seem that his lot is a hard one. They look well dressed and well fed, but very few of them have availed themselves of their opportunities to become the owners of land, while the white laborers, who work side by side with them on the same footing — Germans and Italians — find little difficulty in achieving this, even when they have to pay $30 to $40 an acre for land. In 1890 Washington County, Miss., produced 87,000 bales of cotton, an amount larger than that produced by any county in the cotton belt at that date. This was at the rate of a bale to 1.4 acres. In the same year Tensas Parish, La., made an average of a bale to 1.1 acres, but the aggregate crop only amounted to 40,000 bales, a great falling off from the crop of 1860, which mounted up to 140,000 bales, a yield not yet equaled any- where, and shows to a limited extent — for the country was new in I860 — the capacity of those lands for producing cotton on a large scale under thorough management. The average value per acre of land in farms in this region, as esti- mated in the last census, was $14.48, a rise from the estimated value in 1880, which was $9.70. The increase was greatest in the Eed Eiver country of Texas, where the lands rose from $7 to $17 an acre. The cost of production in 1880 per pound of lint was : Mississippi, 28 estimates, varying from 5 to 9 cents, average 7.4 cents ; Louisiana, 9 estimates, varying from 5£ to 9 cents, average 7.4 cents ; Texas, 5 esti- mates, varying from 3 J to 7 cents, average 5.2 cents; Arkansas, 7 esti- mates, varying from 4 to 10 cents, average 7 cents. The cost of production for 1893 is given as follows : Mississippi, 5 estimates, varying from 6.3 to 11^ cents, average 8.3 cents; 4 esti- mates, excluding rent, vary from 3.9 to 8.2 cents, average 5.5 cents; Louisiana, 3 statements, from 4f to 5| cents, average 4.9 cents; Arkan- sas, 1 estimate, 4 cents. CULTURE OF COTTON. 257 RED-LOAM LANDS. These lauds, comprising in all something over 13,000,000 acres and producing 2.G per cent of the cotton crop in 1890 and a larger percentage (3.3) in 1880, are fouud east and west of the Mississippi River. The largest section occurs in Arkansas, where they rise in concentric terraces from each bank of the Arkansas River to the height of 2,000 feet above sea level and cover over 10,000,000 acres. The cotton crop there was 190,000 bales in 1S90, ah increase of 22 per cent ou the crop of 1S80. It required, however, an increase of 8G per cent in acreage to effect this. One and seven-tenths acres made a bale in 1S80, and 2.6 in 1890. The size of the farms remained the same, averaging 105 acres in each year. Only 31 per cent of the farms were under 50 acres, and the number rented was also- a good deal below the average, but they increased 7 per cent during the eleventh decade. Only 8 per cent of the population was colored in 1880, but increased during the decade to 13 per cent. The number of horses on farms exceeded that of mules in 20 counties, and the mules outnumbered the horses in 2 other counties. Ten estimates of the cost in 1880 of producing a pound of lint, varied from 3.45 cents to 8.5 cents; average, 0.19 cents. The eastern counterpart of the red loam of Arkansas is found in the 2,500,000 acres in the central basin of Tennessee. The cotton crop here fell to 10,504 bales in 1890, from 47,085 bales in 1880, although the acreage only decreased 11 per cent. Ninety-nine per cent of this basin is in farms, a larger proportion than anywhere else in the cotton belt; 61 per cent is in improved land — again a larger proportion than else- where in the cotton belt — and only 3.7 per cent of the area was planted in cotton. Another notable exceptiou to the change occurring else- where is that here we find the only instance where the farms increased in size. A very slight increase, it is true, onl y 106 acres iu 1890 against 104 in 1880, but in line with the average increase of 3 acres which took place during that period in the size of the farms of the United States at large. The number of farms under 50 acres decreased 14 per cent, which is exceptional also; but the number rented obeyed the general tendency, and showed a small increase. It required 2.4 acres to make a bale in 1880 and 4.2 acres in 1890. The population was 37 per cent colored in 1880 and fell to 33 per cent in 1890. Horses were more numerous on the farms in 7 counties, and mules iu 2. THE VALLEY REGION. The Unaka and Cumberland valleys in southeast Tennessee and northwest Georgia, together with the valleys of the Coosa in Alabama and of the Tennessee River in that State and in Tennessee, have been grouped under what is called the valley region, including nearly 1993— No. 33 17 258 THE COTTON PLANT. 15,000,000 acres, much of which is too elevated to be planted in cotton. In 1890 the area in cotton exceeded 795,000 acres, an increase of 41 per cent on the acreage of 1880, showing that cotton growing has made progress in these altitudes, although the product fell 9,000 bales short of what it was in 1880. In that year 4.2 per cent of the entire cotton crop was grown here ; ten years later, but 3.1 per cent. In the first year a bale was made to 2.3 acres, and in the last to 3.4 acres. The larger part of the crop is grown in Alabama. This region is too widely scattered and the importance of its various divisions of too little consequence to the production of cotton to justify a detailed mention of them. It is to be noted, however, that some of the very best staples of upland cotton is produced here. Forty bales raised in northwestern Georgia were quoted in New Orleans and also in Boston, in January, 1895, at 3£ and 3f cents above ordinary uplands of the same grade. THE ALPINE REGIONS. These lands are found in North Carolina, South Carolina, Georgia, Tennessee, Arkansas, and Texas, and produce about 0.9 per cent of the entire cotton crop. The conditions of culture are so various as to preclude general description, and they are not of sufficient importance to require a detailed description. Mention of other cotton-growing localities must also be omitted for similar reasons, and have not been included in any of the general statements that have been given. Among these is Florida, where the largest cotton crop ever made was only 65,000 bales, a crop smaller than that of several counties of the cotton belt; Missouri, that one year grew as much as 20,000 bales; Virginia, that never reached that figure; Illinois, that made 1,482 bales in 1860; Kansas, that made 212 bales in 1890; Kentucky, that made its largest crop of 1,367 bales in 1880; Indiana, that made 14 bales in 1850; New Mexico, that made 19 bales in 1860; Utah, that produced 136 bales in the same year; and California, that made 34 bales in 1870. Of very much greater consequence are the cotton crops of Indian Ter- ritory and Oklahoma, but data are wanting in regard to them. Esti- mates of them were included in the returns made for Texas in 1880 and 1890, and it is very probable that they were underestimates. The crop in 1894 was estimated at over 112,000 bales. The number and wide dis- persion of these localities show the enormous extension that might be given to cotton growing in the United States if there were a profitable margin between the price and the cost of production. GENERAL OBSERVATIONS ON COTTON CULTURE. The matter of the first consideration in the culture of cotton, as in that of any other crop, is to prevent the removal of the soil by washing. Everywhere in the hill country neglect in this regard has resulted in the denudation of the soil from extensive areas of cultivated fields, rendering them barren, and devastating other fields lying at a lower CULTURE OF COTTON. 259 level. Kor does the injury stop here. The public roads become conve- nient channels along which, to their destruction, these muddy floods at last pour into the streams, damming them up, causing freshets, and con- verting fertile bottoms into miasmatic marshes. The evil is generally recognized, and to some, but to a wholly inadequate extent, remedies are applied by terraces and hillside plowing. Where this is thoroughly done, and persisted in, it has proved eminently successful. A very common error has, however, attended the practice. It is that some fall should be given the line of the terrace, to allow the water to escape. The result is, that while one gully may be cured by such a terrace, a larger amount of water is concentrated at its lower terminus, and another and larger gully created there. The terrace should be on an exact level, and must from time to time be amended, on account of changes occurring in the spaces betwen the terraces. A spirit level may be used to establish the line of terrace, but a simpler, cheaper, and more accurate implement is a compass, made of light stuff', and strengthened with a crossbar. The legs should be 15 feet apart at the ground, coming together 7 feet above it. At the apex a cord is sus- pended, with a weight attached to act as a plumb bob. When the feet are on a level, the place where the plumb cord crosses the bar is marked. In stepping off the terrace the level will be exact when the plumb line corresponds with the mark on the crossbar. It would seem proper that legislation should compel owners on the higher levels to restrain the rains, which, falling on their fields, issue in destructive floods on their neighbors' at a lower level.1 DRAINAGE. The usual substitute for drainage in the cotton field is putting the plants on a high bed and cultivating them deep. It is an expensive substitute, and the rice and sugar planters along the border of the cotton illustrate with ample object lessons the benefits and methods of drainage. The trouble is that the original outlay for drainage is con- siderable, while the makeshifts, though in the end more costly, are within the reach of the small cultivator. It would seem desirable for the legis- lature to establish some equitable method of adjusting a right of way for outlet ditches. INCLOSURES. Many strong arguments2 have been brought against fencing in cultivated fields and allowing stock freedom of range, and many laws modifying the old practice have been passed in recent years. Per- haps nowhere has so radical a change been effected in this regard as in South Carolina, where it is required that stock be kept in fenced inclosures and heavy penalties imposed when they trespass. This law has promoted the cultivation of large areas of land. Since its passage 1 U. 8. Dept. of Agr., Farmers' Bui. No. 20. 2 U. S. Dept. of Agr. Rpt. 1871. 260 THE COTTON PLANT. 2,500,000 acres, or 74 per cent more land, have been brought under cultivation, adding largely to the cotton crop, while the grain crops have increased only half as much. In the face of the increased area under culture, the increased crop, the larger population, and the greater wealth, there has been an absolute decrease of 23 per cent of horned cattle, 26 per cent of hogs, and 33 per cent of sheep. It has assisted in the development of a one-sided husbandry, and to the diminution of agricultural values — the value of land among the others. SUBSOILING. Subsoiling and deep breaking are open to question. There is no question that a deep, mellow soil is to be preferred, but the efforts to obtain it are limited by the cost, by the risk of injury to some soils through leaching, and to others by bringing sterile earth to the sur- face. Sandy soils may suffer in the first way, and heavy clays in the second. Experiments to determine the value of these operations are conflicting and inconclusive. ROTATION. Rotation of crops opens a wide field of inquiry. The usual practices have been noticed, and the value of broad leaved and narrow-leaved plants or root crops and crops maturing above ground in rotations might be mentioned, but there is an absence of exact knowledge here, which is a cause of much distrust. The rotation of growths observed everywhere in nature shows its necessity, but this rotation differs with every slight variation of soil, and nothing is accurately known about it. Such knowledge would have to go far beyond the theory of the exhaustion of certain fertilizing constituents of the soil. It would have to deal with hosts of living animal and vegetable friends and foes who fight for or against each growing crop and render changes necessary. , The farmer's mind grows confused over the complicated conditions of this great struggle, and after vainly attempting to understand and con- form to them, he withdraws, turns his fields over to nature, and lets them "rest." And nature, resuming her work of growing heavier and heavier crops every year, restores the fertility which man has destroyed by his exhaustive culture. Even Peter Henderson, the great gardener, said rest was necessary to his gardens once in five years. Exhaustion of the soil differs in intensity, but for the most part it is only temporary. Fields considered utterly used up and thrown out for years when cultivated again have produced better than those which have been under a management more or less careful. JSTevertheless this temporary exhaustion must inevitably occur in every soil not treated to restoratives, notwithstanding that full crops of cotton have been grown on some soils for more than forty successive years. CULTURE OF COTTON. 261 PLANTING AND CULTIVATION. Bedding up land previous to planting is universally practiced. Where manures are drilled in, this is indispensable. It forms a warm seed bed in the cool weather of early spring and possesses other advan- tages. The plants are usually left 2 to 3 inches above the middle of the row, which in 4-foot rows gives a slope of an inch to the foot. This causes the plow in cultivating to lean from the plants, to go deepest in the middle of the row, and, as a consequence, to cut fewer roots. Four feet is the usually accepted distance between the rows. The dis- tance between the plants seems of little importance within the limits of 8 to 14 inches. Still, as nothing but cotton stalks will make cotton, it is unsafe on average land to risk wider spans than 1 foot. ISTothiug conclusive has been settled about checked cotton. It may save a hoe- ing, which should cost about 30 cents au acre, and as plowing is done at about the same cost the question of saving is not determined. The skillful use of the hoe does the most accurate and thorough work. Good crops are made with the hoe without using the plow at all. It may be said that cotton growing was originally established entirely by hoe culture, even the soil for planting being prepared with the hoe. The perfect cotton planter is not yet invented. It should drop five or six seed in a single line at regular intervals, say a foot apart. In very dry seasons a narrow and deep coulter furrow, the dirt closing in behind it, is run immediately in advance of the planter. It freshens up the bed and assists very much the germination of the seed. Much is said about deep and shallow culture, and many believe that they can affect the plant beneficially by practicing the one or the other. The only certainty is that all grass and weeds must be vigorously kept down, and that the capillary pores, through which the moisture escapes after rains, must be broken. The first is most thoroughly effected by a broad, sharp sweep, which takes everything it meets, while going shallower than most other plows. Harrows and cultivators are apt to be turned aside by stubborn bunches of grass, which thus escape them. But the sweep does not distribute the loose dirt as generally as a light harrow does and therefore is not as effective in the mulching process. The effect of cutting roots depends entirely upon the season that fol- lows the operation. The following experiments will show how difficult it is to arrive at results in this matter. In the month of June, the cot- ton plants being 18 inches high, dirt was drawn up G inches around some; it was drawn away to the depth of G inches from others; the roots of others were cut all round close to the stalk to the depth of G inches, and the next stalks had all the roots cut off below G inches. These last wilted in a few moments from this heroic treatment, but neemed to recover in a few days. A rainy season ensued, a vigorous growth set in, and when the crop matured no difference could be observed in the fruitfuluess of the different series of plants. 262 THE COTTON PLANT. The date of cotton planting reaches from March 1 to June 10. Cotton is seldom planted at the latter date, except when put in after a crop of oats. A good crop is made when the season is especially favorable, but the occurrence of drought makes it exceedingly uncertain. The plants also are more liable to the attacks of caterpillars, which only make their appearance in force late in tlie season. They prefer to feed on the younger and fresher stalks, and it was thought in some sections that the frequent recurrence of the cotton worm was in some degree promoted by the late planting of cotton after oats, which was much in vogue at one time. At least they were not so bad after it was aban- doned or before it was commenced. The last regular planting is May 20 under the mountains in Georgia. The following are the dates in the various sections: Planting commences March 1 in southern Texas; March 15, middle Louisiana, Texas coast; March 20, southern Mississippi; March 25, South Carolina coast, pine hills of South Carolina and Georgia, mid- dle Mississippi; April 1, Mississippi bottoms, middle Texas, southern Arkansas; April 5, northwest Georgia; April 7, middle Arkansas; April 10, west Tennessee, Piedmont, North Carolina, South Carolina, Georgia, upper Alabama, north Arkansas, upper Texas; April 20, northern Louisiana; May 20, northeast Georgia. The first blooms appear May 15 in southern Texas; May 20, central Louisiana; May 25, central Texas, southwest Georgia; June 1, Missis- sippi bottoms, southern Arkansas, middle Georgia; June 10, pine hills South Carolina, middle Alabama, central Georgia, Tennessee; June 20, northwest Louisiana, middle Arkansas, northwest Georgia, southern North Carolina; July 4, northern Arkansas, northern Texas; July 10, northeast North Carolina; July 25, northwest Tennessee. The first bolls open May 15 in southern Texas; June 25, middle Texas; July 1, south Louisiana; July 10, middle Louisiana; July 15, southern Georgia, pine hills South Carolina; August 1, nortliwest Louisiana, south Arkansas, coast North Carolina; September 1, Pied- mont, North Carolina, red-loam prairies Texas; September 15, north Arkansas. Picking commences July 10 in southern Texas; August 1, southern Louisiana, central Texas; August 15, pine hills South Carolina, coast of Georgia and South Carolina, Mississippi uplands; August 25, north- west Louisiana, Mississippi bottoms; September 1, north Texas, coast of North Carolina, northwest Georgia; October 1, northwest Texas, north Arkansas. PERIOD OF GROWTH. The following data relating to the above and other important points in the life of the cotton plant are from records carefully kept in South Carolina, near Augusta, Ga. Of 100 seed planted, 10 in a hill, March 29, 1887, 24 came up of which 2 died, 39 could not be found and were probably eaten by insects, 23 rotted, and 14 seemed sound but failed to germinate. CULTURE OF COTTON. 263 The first plant appeared in 14 days after planting; the 10 hills were up to a complete stand in 18 days, and no seed came up after 30 days. This season was cool and wet, but in very dry seasons seed may lie m tlie ground from April 1 to June 10, and then come up to a good stand. The third leaf made its appearance in 8 days after the plant came up and in 22 days after the seed was planted; the fourth leaf appeared the day following. The significance of this observation is that after the true leaves appear, the plant, being no longer dependent on the seed leaves for its supply of nourishment, is not so liable to injury from cold. Other series of seed were put in the ground at later dates and the following observations recorded from day to day: The first form (bud) was seen on a plant coming up in April, 41 days after the plant appeared, and 53 days after it was planted. For all the other plants coming up in April, the average was 40 days to the form, ranging from 34 to 45 days, appearing earlier in the warmer weather and later in the cooler weather. For plants coming up in May, the average was 29 days from the appearance of the plant to the first form, ranging from 25 to 39 days, to which 8 days may be added, to show time from planting to forming. Forms appearing in May bloomed in 21 to 32 days, average 25 d'ays; forms appearing in June bloomed in 20 to 27 days, average 24 days; forms appearing in July bloomed in 20 to 26 days, average 24 days; forms appearing in August bloomed in 21 to 27 days, average 25 days. Blooms appearing in June made open bolls in 45 to 5G days, average 52 days; blooms appearing in July made open bolls in 64 to 71 days, average 65 days; blooms appearing in August made open bolls in 46 to 58 days, average 52 days. Forms on May 24 made open bolls August 9 ; forms on June 24 made open bolls September 21 ; forms on July 24 made open bolls October 8; forms on August 24 made open bolls November 9. As killing frosts occur about November 17, it would seem that the latest blooms that can be counted on would be about September 1. From this it follows that the minimum period from plant- ing to the first open boll is 120 days, and that the maximum period is 157 days. The interval of 37 days between these periods is more than sufficient to fix a full crop of fruit if the condition of the weather is favorable to the plant at the fruiting stage. SHEDDING OF FORMS, BLOOMS, AND BOLLS. When the weather is not favorable at the fruiting stage, the other- wise hardy cotton plant displays its greatest weakness. It sheds its forms, its blooms, and often its half-grown bolls. The following table, condensed from the daily record above referred to, represents to some extent the loss occasioned in this manner. The plants having received very careful attention, the loss exhibited is a good deal below the aver- age sustained in ordinary field culture. 264 THE COTTON PLANT. Proportion of forma which produce holla. Date of coming up April May June July Num- Forms. Blooms Bolls | ber of ami bolls matur- plants. Appeared. Died. Blocmed. dropping. ing. 7 1,700 1,231 4G9 ig:j 30G 10 2,580 1,819 667 199 408 2 154 106 48 12 S6 1 GO 24 36 19 7 Per cent matur- ing. 18 18 23 11 The 1,580 bolls picked before September 10 weighed 20.5 pounds, or about 77 bolls to the pound; 432 bolls, picked September 10, weighed 5.75 pounds, or 75 bolls to the pound; 203 bolls, picked October 5, weighed 3 pounds, or 07 bolls to the pound ; 103 bolls, picked October 21, weighed 3.5 pounds, or 110 bolls to the pound. The average for the whole season was 85 bolls to the pound. These plants were fertilized at the rate of 036 pounds to the acre, one- fourth acid phosphate and three-fourths cotton-seed meal. They were planted in 1-foot rows, 18 inches between the hills, which would give 0,3G0 plants to the acre. If an acre had fruited as these 20 plants did, and every forDi had stuck and matured into an average boll, the yield would have been 25,052 pounds of seed cotton to the acre, a yield undreamed of. As it turned out, they actually produced at the rate of 4,100 pounds to the acre, a yield that has seldom, if ever, been attained. Such calculations show how misleading it is to apply estimates on small patches to field crops, but it also shows that much more might be obtained by greater care and precision. A more thorough study of the cotton plant might discover means to obviate this great waste. At present cotton growers are at a loss to form a correct idea of the cause or to apply any effectual remedy. A week or two before cotton opens, sometimes a month, the crop being clean, field work stops. Formerly much important work of repairs and improvement was done during. this interval. jSTo\v the hands that w^re engaged in cultivating the crop are discharged and no work except what is absolutely indispensable is done. Cotton picking is the most tedious and expensive operation in cotton growing. The picking of the crop of 1801 is estimated to have cost not less than $60,000,000. The most of the picking was paid for at 50 cents per hundredweight, and planters in Texas who grew as much as 2,500 bales said it had cost them $9 a bale to gather their crops. It is very light work, at the most pleasant season of the year, and it is effectively performed by women and even by small children, as well as by. men. In the early days of this culture the amounts of cotton picked were small. It is related in the Southern Cultivator that the report that a young man had picked 100 pounds in a day created great excitement among the farmers in Georgia, who came from far and near to see it CULTURE OF COTTON. 265 done and gave a barbecue in honor of the achievement. For a long time this has been a low average for ordinary pickers. As early as 1839 there is a record of 80 hands — men, women, and chil- dren, old and young — averaging over 133 pounds of seed cotton apiece a day. In October, 1891, 10 convicts of the Mississippi penitentiary picked in 5J days 18,340 pounds of cotton, a daily average of 333 pounds per man. The picking season will average in duration at least 100 days, and picking at the above rate would turn out 22 bales to the hand. It has never been assumed that one man could cultivate more land than would make 10 bales, so that one man is able at average full work to gather as much as two can make. This, however, is very far from being the case, and in the Mississippi bottoms the same year it was not unusual to hear of tenants and space workers who did not gather during the whole season as much as a bale to each picker in their families. Strikes among cotton pickers are not made by combination, but they are exe- cuted as effectually and destructively by individuals. It is very diffi- cult to get them to work until the cotton is fully open, and it is hard to stimulate them to pick over 100 pounds a day. The damage re- sulting from slack work here is often very serious, due iu part to the loss of some cotton by falling out, and to an equal extent to the injury to the quality of that which is gathered, the staple being soiled by dust and stained by the coloring matter from the bolls. It has been thought that the production of cotton would be limited by the amount that could be gathered. This limit is still remote. Excluding the population of towns and villages, who do a considerable share in cotton picking, and deducting one-third for children under 11 years of age, there remains an exclusively rural population in the cotton States of over 6,800,000, all more or less occupied in cotton growing and capable, at the low aver- age of luO pounds a day, of picking daily more than 450,000 bales, or the very large crop of 1894 in 20 or 21 days; and if they did the same task during the whole season they could gather four or five times as .much as the largest crop yet made. Much skill and capital have been expended in efforts to make a machine that would pick cotton. It can not be said that any has proved successful in solving a problem that seems about equal to that of gathering strawberries or raspberries by machinery. This could be done if it were not for the injury to the berries; and if they were to be made into jam, perhaps assorting and washing machines might be invented to utilize a portion of the harvest. Cotton-picking machines gather limbs, leaves, and bolls, and pass the whole through a cleaning separator that, it is claimed, leaves the cotton in the condition of aver- age cotton picked by hand. A cotton-picking machine with a driver and two horses, taking a row at a time, would go over about G acres a day. The cost of the work of an expensive and complicated machine, as this must necessarily be, would hardly be less than $5 a day, and if the cotton were Gathered at the right staa;e there should not be more 266 THE COTTON PLANT. than 200 pounds to the acre open. Cotton left in the field for a fuller opening than that is liable to serious damage, and in case of storms to almost total loss. The machine would thus gather at the most 1,200 pounds a day at a cost of 41.G cents per hundredweight, the present cost being from 40 to 50 cents, and so highly paid at that that there is little doubt it will be reduced to 30 cents or less (as it has been already in some localities), and even then expert pickers will earn from $1 to $1.50 a day and more. In improving short staple cotton there is a growing tendency to develop varieties which take on and opeir all their fruit at nearly the same time. If such a variety were perfected, it would simplify the gath- ering by machinery, especially as such varieties at present shed their leaves about the time the cotton begins to open, thus removing the character of trash which it is more difficult to separate than either the stems or the bolls. COST OF COTTON PRODUCTION. The cotton crop that it is possible to grow and gather has been shown to have no practical limits, either in the area of land on which it can be cultivated or in the labor available to gather it. It may be added that one-fifth of the horses and mules at present on farms in the cottou States would suffice to cultivate the largest crop that has ever been put in. More laud can always be prepared and planted by a given amount of horsepower than can be cultivated by it. The practical limit is the amount of land a work animal can cultivate between May 5 and August 1. From the 87 days of this period something over 12 must be deducted for Sundays, and as there is an average of 28 days' rain at this season in the cotton belt on which no plowing can be done, deduction must be made for these also, leaving 40 working days. Four plowings, sometimes more, must be given in good cultivation, which would leave Hi days for each plowing. A day's work being 3 acres, the horse would work 34i acres. By following out this calculation a good idea of the cost of production under present conditions of cotton culture may be obtained. To prepare the land it should be broken up at the rate of an acre a day, but as this is rarely done more than once in three years, one- third only is to be charged, say, 11 days' work on this account; the laying off, bedding, dragging off, and planting will require a little over nine-tenths of a day's work per acre, or 33^- days' work in all to prepare and plant, making a total before cultivation commences of 44£ days' work. To this, if Sundays are added and rainy days as before, it will sum up to 07 days, and adding the 87 days for cultivation, the total is 154 days' work of a horse to cultivate a full crop of cotton. But it remains to haul the crop to the gin. Estimating the crop at the average crop of the cotton belt for the last eleven years, viz, 531 pounds of seed cotton to the acre, the mule would have to move 18,319 pounds from the 34£ acres; at four loads of 800 pounds a day there would be 0 days' work; in all, 160 days in a year. The cost of a horse, counting insurance, feeding, stabling, and CULTURE OF COTTON. 267 interest, together with the wear and tear of such gear and implements as are used in the cultivation of cotton, is fully covered by $122.85 per annum. For the portion of the year the animal is engaged in the cotton crop this would be 853.92. The wage of the plow hand for the crop niaj* be put at $61. Four hoeings would cost for the 31i acres $55.20, making the total cost to date of picking and cultivating 34J acres in cotton 8173.12. Picking the amount stated above, as an aver- age crop, would cost $73.27. In all, the cost for delivering at the gin- house 18,319 pounds of seed cotton would be $216.39, or 1.314 cents per pound. If fertilizers are used at the rate they were used in 1890 in Georgia and South Carolina, $1.78 an acre must be added on that account, bringing it up to $307.80, or 1.68 cents per pound of seed cotton. It remains to mention rent, taxes, and management. As the latter consists almost wholly in seeing that advances are charged against the laborers and in collecting such due as it is not difficult to get, 5 per cent commissions on the profits would be a fair allowance. Taxes are so inconsiderable that they do not merit notice. With rent it is dif- ferent; economic rent does not exist in this country; the most usual rent has been a portion of the crop, and one-fifth to one-fourth of the cotton produced has been paid. Xevertheless rents have differed a good deal; in the east $2 to $3 has been a fair average; in the west, in the Mississippi bottoms and the black prairies it has run from $1 to $10 an acre. The low prices of cotton have changed this, and in some sections, even of the black prairies of Texas, tenants have been notified by merchants that they could not advance to them in 1895 if they paid a money rent exceediug$3 an acre. It is safe to say that this year an average rent is $1.50 an acre in the east and $3 in the west. The first amount is to be added to the cost of production where fertilizers are used, and the latter where they are not used. This addition makes the cost of a 1-horse crop exclusively in cotton in the east sum up to $359.55, or to 1.96 cents per pound of seed cotton; in tlie west the result will be $319.89 for a similar crop, or 1.88 cents per pound of seed cotton. This would be very nearly a cost of 5.88 cents per pound of lint in the east and 5.68 cents in tbe west for cotton delivered at the ginhouse, the cost of ginning and packing remaining to be counted, the value of the seed more than sufficing for this in both instances. The seed commands a higher price in the east, and would tend by its sale to render the cost more equal. Though worked out by a different method, these results will be found to differ little from an average of tbe various estimates already given.1 Tbe percentage of tbe cost of labor on tbe total cost of production is 53 per cent in the east and 55 per cent in the west. This is a much higher ratio of the cost of labor than will be found in 'It should be clearly understood that none of these estimates do more than indi- cate the probable cost of cotton production, assuming certain prices and conditions which may or may not obtain iu any given case. They may approximate the average cost of production under present systems of cotton culture in the South, but they by no means show the minimum of cost under the most improved methods. — Ed. 268 THE COTTON PLANT. most other industries. In averaging the percentage of the cost of labor to total cost in 10 cotton mills taken at random from among the Northern mills, and the same number from the Southern, the first is 33 per cent and the second 22 per cent. Labor is thus seen to be the pre- dominant element in cotton growing. Modern improvements in machinery, which have so wonderfully affected most industries, have only indirectly benefited agriculture, and cotton culture as little as any branch of this pursuit. Great im- provements have been made iu agricultural machinery, but the share worker and small tenant, even the one-horse farmers, whose numbers are rapidly multiplying in the cotton-growing region, are not able to avail themselves of such helps. The improved implements are as a rule the property of another class, and the profit from their use goes almost exclusively to their owners. It costs as much in human labor to prepare, plow, hoe, and pick an acre of cotton now as it did half a century ago, and if the product is increased by the successful use of commercial fertilizers, such use necessitates an increase of labor in the preparation for and the culture and gathering of the product. On the other hand, the agricultural laborer finds in strikes no resource to benefit him. The seasons inex- orably fix the day and hour when his services are needed ; if he allows it to escape him, it is gone once for all, and with it disappears the product out of which alone wages are paid. The employer also can not resort to lockouts. The manufacturer may suspend operations and lose little besides the interest on his plant. Such a suspension even for a brief period of farm work would mean the loss of the entire season's work. The small farmers, working on a narrow margin, are always in imminent need of cash, and cotton is the only crop that never fails of a ready cash sale. Every pound of it can be readily disposed of by the producer for cash, and at the prices quoted in the mar- kets of the world. All other crops, unless grown upon a scale suit- able for shipment in bulk — a scale seldom within the reach of the small farmer — are subject to the vicissitudes of the local market, easily overstocked, and often inflicting heavy loss on the producer of perish- able commodities. The general testimony is that, while farmers growing cotton exclu- sively are in very bad condition financially, those who raise food and for- age crops, and especially those who in addition raise their work animals, are everywhere prosperous. The chief reason of this is, as has been shown, that the exclusive cotton grower fails to employ the most important forces in farm work, the work animals and the land, to the fullest extent. He draws upon the surplus of a single crop made in part of the year for the means to support his farm during the whole year. There is a saving in using unemployed time and capital to produce necessaries which otherwise must be paid for in money. The tendency to pay greater attention to food and forage crops has been much accelerated by the low price of cotton since 1890. CULTUEE OF COTTON. 269 COTTOX PRODUCTION IN DIFFERENT STATES. The usual method pursued in collecting- and arranging the data of cotton production has been to do this by States. The attempt has been made in this article to marshal these data in what seemed to be the important natural subdivisions or regions of the cotton belt without confining them within State lines. These natural subdivisions are sub- stantially those adopted by Professor Hilgard in his work on cotton culture for the Tenth United States Census. It was thought that these regions of the cotton belt possessed well-marked characteristics of soil, growth, products, climate, and methods of culture, together with racial, social, and industrial peculiarities, all of which were influenced only in a very small degree by the political boundaries of the States. There- fore it would seem that the understanding of these real and natural regions is more important to a knowledge of the conditions of cotton culture and production than the artificial political framework of the States. But as the production of cotton by States maybe of inteiest, the following data are given: Percentage of cotton crop grotvn in eacli State. State. 1799. 1809. 1819. 1829. 1839. 1849. 1859. 1869. 1879. 1889. 1S90. 1891. 1892. 1893. 1894. Texas Georgia 27. Mississippi Alabama Arkansas South Carolina 44. Louisiana North Carolina 11. Tennessee 3. Florida Others 13. 25.0 6.2 12.5 50.0 2.5 8.5 4.0 10.0 27.5 6.3 6.2 11.3 5.0 100.0 100.0 100.0 21.4 13.3 14 1, 18.3 12.4 4.7 12.0 2.0 20.6 24.6 14.7 0.7 7.8 19.3 6.5 3.1 1.5 1.2 2.4 20 2 19.6 22.8 2.7 12.2 7.3 2.9 7.9 1.9 0.1 8. 01 13.0 22. 41 18.5: 6 8| 6.5| 14.5 2 7 5.5 ' 1.2 0.9 11.6 15. 6: 18.8' 14.3 8.2; 7. 3J 11.6 4.6| 6.0 1.3 0.7l 14.2 14.2 16.7 12.3 10.5 9 1 8 8 6.8 5.8 0.9 0.7 19.8 16.0 15.8 12.2 9.3 10.0 8.9 4.5 2.5 0.7 0.3 23.0 14.6 14.6 12.2 9.6 8.8 7.3 4.3 2.5 0.6 3.1 100.0 100.0 100.0 100.0 100.0 26.8 13.4 13.9 11.9 8.8 8.7 7.0 5.4 3.6 0.3 0.2 31.1 14.0 11.3 11.0 7.9 9.4 5.0 5.5 3.9 0.6 0.3 27.3 33.1 14.9; 13.2 12.1 12.1 12.3 10.1 8. 4 8. 6 9.9 5.4 5.6 3.4 0.6 0.1 8.0 6.0 4.7 3.5 0.6 0.1 100. 0 100. 0 The above table presents a general view of the order of development of cotton growing in the different States. It shows the extraordinary expansion of the industry in the extreme West, but the success with which the Eastern States have kept up their ratio of percentage to the whole crop for half a century demonstrates the deep root this culture has taken in those States. Average annual yield per acre of cotton in different States. Pounds of lint cotton per acre. 1 a 218 215 206 208 252 373 383 335 229 254 ca a '3 140 134 122 110 150 182 142 124 126 139 6 CD State. -£ i to OO 00 C5 O L- OO OO OO OO OO OO GO 175 '134 '122 120 186 254 244 221 194 157 -* o OO OO OO OO 174 176 138 148 124 146 '110 125 170 175 243 224 '142 175 193 209 174 172 150 165 CD OO OO 157 140 137 130 178 221 200 240 171 162 t~ OO OO 196 168 158 135 196 228 163 233 168 T77 OO OO OO 202 155 159 136 192 206 lfi-l OO OO '140 180 170 165 200 252 183 o OO 2218 2215 197 165 230 293 233 290 198 220 o OO 194 204 180 179 2 252 2 373 260 2 335 225 247 < N. Carolina.. S. Carolina. . . Georgia Alabama Mississippi .. Louisiana 153 151 137 144 124 136 111 135 '150 185 199 195 150 204 '124 200 133 154 '139 171 195 198 172 185 140 163 136 154 202 172 263 ' 182 167 230 263 215 204 203 177. 188 180 183 152 150 190 235 240 243 170 200 178 162 155 146 192 249 O90 Arkansas Tennessee . . . Cotton belt . . 230 169 169 201 U26 172 244 172 183 1 Minimum yield. 2 Maximum yield. 270 THE COTTON PLANT. This table is an approximate estimate of the product, in pounds, of lint cotton per acre of each State.1 For the first eight years recorded the average yield was 1G8 pounds lint per acre; for the ninth year it was 177 pounds, and for the eight succeeding years 204 pounds of lint per acre. The improvement seems to have been very uniform in all the States. The maximum yields have been made in the last five years and the minimum yields in the years previous. It goes to show that there is a steady and progressive improvement in the culture of cot- ton which must distance foreign competition. It Avill be noticed that the difference between the maximum and minimum yields is greatest in the States making the greatest yields, which are also those States that have entered most recently on cotton growing. The fluctuations are largely due to the variations in the seasons. Thus, in Xorth Caro- lina the minimum and maximum yields stand side by side in the years 1889 and 1890. It seems, however, by no means probable that the steady increase in the yield per acre is in any considerable degree due to more favorable seasons. The following table gives the United States Department of Agricul- ture estimates of the total production of cotton in the different cotton- producing States during the years 1894 and 1895 : Cotton crops of 1S94 and 1895, by Stales and Territories. States and Territories. Acres. Bales. Bales per acre. 1894. 1895. 1894. 1895. 1894. 1895. 2. 664, 861 1, 483, 319 201, 621 3, 610, 968 233, 898 168 8,243 1,313,296 2, 826, 272 63, 696 1, 296, 522 28, 992 2, 160, 391 879, 954 6, 854, 621 2,371,726 1, 186, 655 191,540 3, 069, 323 212, 847 40 1, 142, 568 2,487,119 47, 772 1,050,183 26, 093 1, 814, 728 712,763 5, 826, 428 400 44, 623 854, 122 709, 722 48, 005 1, 183, 924 104, 887 67 2,685 721,591 1, 167, 881 24, 114 454, 920 13, 001 818, 330 286, 630 3, 073, 821 663, 916 520, 860 38, 722 1, 067, 377 68, 668 15 0. 32 .48 .24 .33 .45 .40 .33 .55 .41 .38 .35 .45 .38 .33 .45 .21 0.28 .44 .20 .35 .32 .38 513, 843 1, 013, 358 11, 816 397, 752 14, 103 764, 700 172, 560 1, 905, 337 103 7,964 .45 .41 .25 .38 .54 South Carolina .42 .24 .33 Utah .26 61, 128 12, 735 .18 Total 23, 687, 950 20,184,808 9,476,435 7,161,094 .40 .35 Note. — The figures given in this table were secured by a method adopted by this Department in September, 1894, with the view to giving the planter reliable information as to the crop of the year before the beginning of the following season. The cooperation of all railway and water transporta- tion companies and cotton mills operating in the cotton States, as well as all custom-houses and several thousand cotton buyers, merchants, and ginners in those States, was secured in reporting data relating to the movement and consumption of cotton and interior stocks on hand. The estimates from data thus secured are made as follows: The movement from all interior points in each State across itsboundaries is carefully investigated, and themovementto the ports of a State is included with the movement from the State, because practically all receipts at ports in the cotton section are ultimately shipped coastwise to Eastern points or exported. Prom the reserve stocks of the State the cotton at its ports is therefore excluded, being already accounted for in the railway and water movement. Further deduction is necessarily made of cotton shipped from one cotton State to interior points in another. In the season 1894-95 the movement was followed from September 1, 1894, to April 1, 1895 ; in 1895-96 it was followed during the entire season, from September 1, 1895, to August 31, 1896. ]The data are obtained from the United States Census returns for the years 1879 and 1889, from the reports of the United States Department of Agriculture, from Latham, Alexander & Co.'s Cotton Movement, and from Shepperson'o Cotton Facts. CULTURE OF COTTON. 271 EXPERIMENTS IN COTTON CULTURE BY THE EXPERIMENT STATIONS. The agricultural experiment stations in the cotton belt have con- ducted relatively few held experiments on cotton culture, their work in this line having been largely confined to the best distance between the plants and the effects of topping. While the results of these experi- ments have not been conclusive, some of them afford suggestions of value to the planter and the experimenter. The following brief sum- mary, prepared by Mr. J. F. Duggar, of this office, presents an outline of the work of the stations in this line, with references to the published reports : In a single test at the Alabama Canebrake Station, planting on high beds resulted in a larger yield than was obtained with flat beds on a black slough bottom, a kind of soil which is very retentive of water. There was only a very slight difference in yield between ridges made on an unbroken center and those formed by bedding on a center furrow. The usual recommendation of the stations is to plant on low or flat- tened ridges.1 At Camden, Ark., on a field previously planted in corn, a larger yield was obtained by breaking the land and then forming the beds than by making the ridges without previous plowing, the difference being 292 pounds of seed cotton per acre. On ridges made in February the yield was slightly greater than on those made just before planting in May.2 Subsoiling proved profitable in one test made at Athens, Ga.; and in the same locality results favored planting later than April 10.3 At the Alabama Canebrake Station cotton growing on tile-drained land was in most seasons more productive than on undrained land.4 The depth and extent of the root growth of a plant furnish sugges- tions as to methods of preparing and tilling the soil. At the South Carolina Station the taproot of the cotton plant extended to a depth of more than 3 feet when the plant was grown on a light sandy soil, with subsoil of the same character. On a loam with a more compact sub- soil the taproot terminated abruptly as soon as the hardpan was encoun- tered at a depth of 9 inches below the surface. At the Alabama College Station the taproot penetrated vertically to a depth of 12 inches,. or 3 inches into the subsoil, its course then becoming horizontal. In South Carolina it was observed that most of the lateral roots commenced about 3 inches below the surface, and never went below the upper 9 inches of soil. 1 Alabama Canebrake Sta. Bui. 4. -Arkansas Sta. Bui. 28. 3Proc. Georgia State Agl. Soc, Feb. 1874. p. 67. 4 Alabama College Sta. Bui. 4 (1887), Bui. 3 (1888); Alabama Canebrake Sta. Buls. 11 and 14. 272 THE COTTON PLANT. In a garden soil of sandy drift and pebbles the Alabama Station found a young cotton plant 3{ inches high having one of its lateral roots 3 feet 4 inches long, the end of the root being only 3 inches below the surface. Almost identical measurements were made of the roots of a young cotton plant at the Arkansas Station. In Alabama, from a cotton plant 2 feet high and just beginning to bloom, one lateral root extended more than 5 feet. Some of the lateral roots began only 1J or 2 inches below the surface. The position of the. roots was such that the experimenter estimated that the usual deep cultivation would have destroyed four-fifths of the lateral roots which extended at right angles to the row. In this we have a strong hint as to the superiority of shallow over deep cultivation, a superiority which was proved by experiments ex- tending over several years at the Alabama College, Alabama Cane- brake, Georgia, and Mississippi stations. We find only two instances in which shallow culture failed to afford a larger yield than deep culture.1 At the Georgia Station an experiment to determine the best distance between cotton plants was conducted in five different years. The rows were uniformly 4 feet wide, and the attempt was made to leave single plants either 1, 2, 3, or 4 feet apart in the drill. It generally happened, however, that the stand was much more imperfect on the plats planted close than on the others. This is equivalent to saying that the least average distance between plants was somewhat greater than 12 inches. The following table gives the yields obtained each year with different distances, on land heavily fertilized: Yield in pounds per acre of seed cotton from planting at different distances. Tear. 1891 1892 1893 1894 1895 Average of 5 years 1 by 4 feet. l^ounds. 1,943 1,616 1,903 2,065 2,270 1,960 2 by 4 feet. ■ 3 by 4 feet. I Pounds. 2, 027 1,516 1,905 1,812 2,047 1,861 Powids. 2,007 1,501 1,925 1,843 1,985 1,852 4 by 4 feet. Pounds. 1,833 1,439 1,770 1,671 1,767 1,696 The figures giving the average yield for five years indicate that even on land so rich or well fertilized as to produce 1^ bales of cotton per acre, a distance of 4 feet in the row between plants reduces the yield considerably. On the whole the results seem to indicate that with 4-foot rows, when the date of planting is rather late, there is an advan- 1 Alabama Canebrake Sta. Bui. 4; Alabama Dept. of Agr. Bui. 6 (1886); Alabama College Sta. Bui. 4 (1887), Bui. 3 (1888); Arkansas Sta. Rpt. 1888, p. 117; Georgia Sta. Buls. 11 and 16; Mississippi Sta. Rpt. 1889, p. 13, Rpt. 1890, p. 16; South Caro- lina Sta. Rpt. 1889, p. 84. CULTURE OF COTTON. 273 tage in spacing cotton plants 12 to 16 inches rather than in allowing more room in the drill. The experimenter, however, in summing up the results for the five years, expresses a preference for a distance of 2 by 4 feet when early planting is practicable and when a yield of 1,800 to 2,000 pounds of seed cotton can be expected. In this case early thinning and rapid and thorough cultivation are recommended. That close planting favors early maturity, and hence is desirable when the date of planting is late or where the growing season is short, as in the northern part of the cotton belt, is indicated by the following facts noted in the experiment of 1892 : The total yield varied little for plantings at different distances, but at the first picking the yields for distances of 1, 2, 3, and 4 feet were 593, 449, 323, and 221 pounds, respectively. At the second picking the yield was again greatest with close planting. However, at the third picking the yield was greater as the distance was greater, and this was yet more strikingly true at the fourth picking, showing the tendency of wide spacing to delay maturity. The yield of seed cotton per plant as influenced by distance in the drill was determined in the same experiments, as follows: Yield of seed cotton per plant at different distances in rows 4 feet apart. Tear. 1 foot. 2 feet. • 3 feet. 4 feet. 1801 Pound. 0.210 .169 .275 .243 .275 Pound. 0. 405 .287 .417 .325 .411 Pound. 0.565 .423 .561 .483 .557 Pound. 0.687 1892 .542 1893 .690 1894 .566 1895 .665 .234 .309 .518 .626 From this table we see that with plants at a little more than 12 inches apart each plant averaged more than one-fifth pound of seed cotton; at 2 feet apart, a little more than one-third pound; at 3 feet apart, one- half pound; aud at 4 feet, five-eighths pound. The yield per plant varied greatly in different years. In this connection it should be remembered that the minimum yield in these experiments was 1 bale per acre. In 1893 the Georgia Station conducted an experiment to ascertain the best distance between the rows. Each plant was allowed G square feet of ground, one series of plats bearing plants at distances of 3 feet by 24 inches, another 4 feet by 18 inches, another 5 feet by 14.4 inches, and another 6 feet by 12 inches. The yield was greatest — 1,964 pounds of seed cotton x>er acre — when the distance was 3 feet by 24 inches, and the yield per acre and the product per plant decreased as the rows were widened, with the accompanying closer planting in the drill. In other words, the yield increased as the space assigned to each plant 1993— No. 33 18 274 THE COTTON PLANT. approached a perfect square. A repetition of this test in 1895 con- firmed the results of 1893. " On land of less capacity than 1 bale per acre it would probably be well to reduce the width of rows to 3£ or even 3 feet. It may be safely urged that land which will not produce the maximum crop of which it is capable with rows not less than 3 feet wide can not profitably be cultivated in cotton."1 In a test made by the University of Georgia, at Athens, the yield decreased very slightly and very gradually as the rows widened from 2J to 4 feet, the difference being scarcely sufficient to pay for the extra expense of cultivating narrow rows. In this experiment the plants stood close together in the row — 10 to 15 inches apart — and the yield averaged a little more than half a bale per acre. On the same field single stalks and groups of 2 and 3 plants were left in a place, the interval for all being the same — 10 to 15 inches. The yields were prac- tically identical.2 Other experiments, notably those conducted at the North Louisiana Station, suggest that the cotton plant, under some conditions, does not surfer from the presence of 2 stalks in a place, a point of advantage where planting in checks is desirable. At the North Louisiana Station it was found in 1888 that with close planting in the drill every increase in the width of the row beyond 4 feet reduced the yield. As between narrower rows the results were not conclusive.3 In 1889 single stalks were left 8, 12, 16, and 20 inches apart in the drill; 2 stalks in a place were also left at these distances and at 24 inches apart. The maximum yield of seed cotton was 1£ bales per acre. Whenever the distance was greater than 16 inches, the yield decreased whether 1 or 2 plants stood in a place. Even closer plant- ing seemed to be slightly advantageous.4 The above-mentioned experi- ment was repeated in the following year, when the maximum yield of seed cotton — 1,690 pounds — was obtained on the rows in which 2 plants stood together at intervals of 16 inches. For distances greater than 16 inches the yield decreased as in the precediug year, both for 1 and 2 stalks in a place. With single stalks the yields at 8, 12, and 16 inches were practically identical. Planting at a distance of 16 inches has the advantage of rendering cultivation easier than closer planting.5 When again repeated in 1891 the maximum yield — 1,800 pounds of seed cotton per acre — was obtained on the 2 plats having 2 stalks in a place at intervals of both 16 and 24 inches. For single stalks distances of 16 and 20 inches proved of equal value, and yielded more than closer plant" ing.6 The next year the maximum yield — 960 pounds of seed cotton per acre — was obtained from 2 stalks in a place at intervals of 24 inches. 1 Georgia Sta. Buls. 11, lfi, 20, 24, and 27. 2Proc. Georgia State Agl. Soc, Feb. ,1874, p. 67. 3 Louisiana Stas. Bui. 22 (old ser.). ••Louisiana Stas. Bui. 27 (old ser.). 5 Louisiana Stas. Bui. 8 (2d ser.). 6 Louisiana Stas. Bui. 16 (2d ser.). CULTURE OF COTTON. 275 Single stalks were most productive at distances of 1G and 20 inches, yielding, respectively, 010 and 020 pounds of seed cotton.1 In 1803 the greatest yield — 1,160 pounds of seed cotton per acre — was produced by 2 stalks together at a distance of 24 inches. Single stalks at intervals of 12 inches gave a yield almost identical with the above — 1,140 pounds per acre.2 In 1802 early maturity was apparently favored by rather close plant- ing; at the first picking the single stalks or groups of two growing at intervals of 12 inches had matured a larger proportion of their total crop than had the plants grown at any other distance. In 1803, how- ever, close planting did not notably increase the proportion of total yield secured at the first picking. In 1803 at the experiment station at Baton Rouge, La., 1 and 2 stalks were left at intervals of 12, 18, and 24 inches in rows 3, 4, and 5 feet apart. In rows 3 feet and 4 feet apart, a distance of 18 inches between single plants afforded the largest yield; in rows 5 feet apart, 12 inches between plants in the drill proved best. The greatest yield for single stalks, 2,037 pounds of seed cotton per acre, was obtained when the distance was 18 inches by 3 feet; in other words, when the feeding area of each plant approached nearest a square. When 2 stalks were left in a place the greatest yield on any plat was obtained by planting at distances of 2 by 3 feet; in 4-foot rows, 18 inches gave the largest yield, 1,734 pounds; in 5-foot rows the yield increased with the distance between the groups of plants. Taking the average results of all distances with both 1 and 2 stalks in a place, the yield of seed cotton per acre was 1,821 pounds on the 3-foot rows, 1,557 pounds on the 4-foot rows, and 1,540 pounds on the 5-foot rows. With single stalks 3-foot rows and 4-foot rows afforded practically identical yields when the distance between plants was 1 foot; at greater distances in the drill 3-foot rows proved superior. With 2 stalks in a place 4-foot rows proved most productive except when the distance in the drill was extended to 2 feet, when 3-foot rows gave the largest yield obtained on any plat. In this experiment distance of planting did not notably affect the earliuess of the crop.3 The South Carolina Station conducted experiments on the subject extending over several years at three different localities in the State. The average results showed only very slight differences in yield, whether the rows were 3 J, 4, or 4i feet apart. In the South Carolina tests when checking was practiced there were no constant differences in yield, whether the distance was 2i or 4 feet between the hills. In comparing drill culture with checking the aver- age results of a number of experiments were quite similar, indicating no marked differences in yield between the two systems.4 At the North 1 Louisiana Stas. Bui. 22 (2d ser.). 2 Louisiana Stas. Bui. 29 (2d ser.). 3 Louisiana Stas. Bui. 28 (2d ser.). 4 South Carolina Sta. Rpt. 1888, p. 274; Rpt. 1889, p. 324; Bui. 2 (n. ser.) ; see also Georgia Sta. Bui. 11. 276 THE COTTON PLANT. Carolina Station, in 1886, checks 2 feet by 3 feet and 3 feet by 3 feet afforded practically the same yields — nearly 1£ bales per acre — but planting at distances of 4 feet by 4 feet greatly reduced the yield. At distances of 2 by 3 feet the crop matured somewhat earlier than on plats where more space was allowed each plant.1 At the Alabama College Station cotton was planted in 188G at inter- vals of 1, 2, 3, and 4 feet, in 4-foot rows. The closest planting yielded slightly more, the widest spacing somewhat less, than did the inter- mediate distances. The yields at different distances ranged between 1,085 and 1,292.5 pounds of seed cotton per acre.2 Distance experiments were also made in 1889 and in 1890 without decisive result, the figures in the latter year suggesting a slight supe- riority of a distance of 2 by 4 feet as against 1 by 4 feet, 3 by 4 feet, and 4 by 4 feet, the yield being at the rate of about two-thirds of a bale per acre. Close planting hastened maturity.3 In 1891 Welborn Pet, a cluster variety, planted in rows 4 feet apart, yielded 2,519 pounds per acre of seed cotton when the interval between plants was 1 foot, 2,010 pounds when the plants were 2 feet apart, 2,077 pounds at 3 feet, and only 1,145 when the distance was increased to 4 feet. Peeler, a long-limbed variety, in 4-foot rows yielded at 2 feet 1 ,983 pounds, at 3 feet 1,487 pounds, at 4 feet 1,453 pounds, and at 5 feet 1,333 pounds of seed cotton per acre.4 At the Alabama Canebrake Station, on slough bottom land, a dis- tance of 3 by 4 feet resulted in a yield of 952 pounds of seed cotton per acre, against a crop of 896 pounds obtained on the plats where the spacing was 1 by 4 feet, 2 by 4 feet, and 4 by 4 feet.5 At the same station in the following year the yield per acre of seed cotton in 4-foot rows was as follows : Plants 1 foot apart, 1,216 pounds; 2 feet, 936 pounds; 3 feet, 760 pounds; 4 feet, 880 pounds.6 Prom the results just summarized it appears that there has been found no definite law determining the proper distance applicable to all conditions. The results thus far attained by the stations relative to distance between cotton plants, though not capable of generalization, afford useful hints to cotton growers whose soils resemble those of the different experiment stations. Future investigations may result in rules of practice applicable to certain characters of soil and varieties of cotton. At best, however, variations in weather, which can not be foreseen, will limit the application of such rules. Rotation experiments have for some years been in progress at the Arkansas and Louisiana stations, but in the nature of the case 1 North Carolina Sta. Ept. 1887, p. 127. 2 Alabama College Sta. Bui. 4 (1887). 3 Alabama College Sta. Buls. 4 and 22. 4 Alabama College Sta. Bui. 33. 5 Alabama Canebrake Sta. Bui. 3. 6 Alabama Canebrake Sta. Bui. 4. CULTURE OF COTTON. 277 conclusive results are not to be expected until many crops have been grown.1 Five hundred bolls of the Peerless variety were selected at the Arkan- sas Station, from the bottom and from tbe top of well-developed stalks, the seed cotton from 500 bottom bolls weighing 8.20 pounds, from 500 top bolls G.4G pounds. On planting the seed of these two lots of cotton, tbe seeds from bottom bolls germinated much better and more promptly than those from top bolls. The former also matured earlier and afforded a larger yield — 1,043 pounds of seed cotton per acre, as against 700 pounds from seed obtained from top bolls. The more complete and earlier sprouting of the seed from bottom bolls may have been the cause of the earlier maturity and greater productiveness of tbe resulting plants, for two replantings of seed from top bolls were necessary, and even then only about half a stand was obtained, while seed from bottom bolls afforded an excellent stand without replanting. But it does not appear whether the position on the parent plant, the greater size of bottom boll seed, or some other condition was the cause of the earlier and more complete germination of the seed from bottom bolls. Further experiments are needed to fully establish the difference, if any, in the germination and productiveness of seed from different parts of the cot- ton plant.2 Experiments intended to ascertain the effect of topping cotton have been conducted by the Alabama Canebrake, Alabama College, Georgia, Louisiana, Mississippi, and South Carolina stations, and by the Univer- sity of Georgia. In only one of these experiments, that of the Ala- bama Canebrake Station, were the results decisively in favor of topping. In one year of the test at the Alabama College Station the figures slightly favored topping, though in the preceding year the slight advantage was with the plants not topped. The Georgia Station, in 1890 and 1891, obtained a smaller yield from topped plants than from those not topped. Here, as in several other experiments bearing on this question, the effect of topping at different dates was studied, and the single exception to the injurious influence of this mutilation occurred in 1891 on the plat where it took place late, August 15. The earliei the topping the greater was the injury in these experiments. At the Mississippi Station topping as late as September 20 resulted in a large shrinkage in yield. The South Carolina Station conducted experiments in two localities in that State during three years without being able to observe any perceptible variation in yield between plants topped and those not topped. A single test at the iSorth Louisiana Station, in which plants grow- ing at different distances were topped, revealed no marked effect for good or ill resulting from this practice. 'Arkansas Sta. Bui. 23; Louisiana Bui. 17 (n. ser.). 2 Arkansas Sta. Bui. 23. 278 THE COTTON PLANT. Topping lias given contradictory results under different conditions. Differences in soil and climate are probably responsible for this, and it remains for future experiments to determine the conditions under which topping is beneficial or otherwise. In a single test made by the University of Georgia the yield on the plats where topping was practiced was 1,303 pounds of seed cotton and on the untreated area 1,387 pounds; the loss of 84 pounds was ascribed to topping. The prominent feature of this experiment was the probable influence of topping in hastening maturity, the topped plants yielding 47 per cent more at the first picking than did the plants not topped. This increased earliness, if it could be fully estab- lished, might explain the occasional success of topping, and would rec- ommend the practice for late varieties and for localities where the growing season is short. However, two experiments on a more exten- sive scale at the Georgia Station fail to show any perceptible increase in earliness as the result of topping. In future experiments in topping cotton it is to be hoped that its possible influence on early maturity of the plant will be observed. The detailed results of experiments in top- ping cotton may be found in the following publications: Alabama College Sta. Bui. 4 (1887), Bui. 4 (1888) ; Alabama Canebrake Sta. Bui. 4; Proc. Georgia State Agl. Soc. Feb., 1874, p. 67; Georgia Sta. Buls. 11 and 16; Loui- siana Stas. Bui. 27 (old ser.) ; Mississippi Sta. Rpt. 1889, p. 13; Soutb Carolina Exptl. Farm Rpt. 1883-1886, p. 33; South Caroliua Sta. Rpts. 1888, p. 281, 1889, p. 332, and Bui. 2 (n. ser.). DISEASES OF COTTON. By George F. Atkinson, M. S., Professor of Botany in Cornell University. GENERAL NATURE OF COTTON DISEASES. Investigations, continued for several years, have brought to light sev- eral quite well characterized maladies of the cotton plant in the United States. Some of these are physiological in their nature, being due to disturbances of nutrition and assimilation. Other diseases of this plant are due to the action of fungus organ- isms, which live as parasites in various parts of the plant, consuming the nutriment and causing destructive changes, which bring about the death of the part attacked if not of the entire plant. The term " rust," frequently defined as " red rust" or "black rust," has become so general in its application as to be utterly valueless other than in conveying the notion of disease. If we accept the term " cotton rust" as simply syn- onymous with cotton disease, it will tend to eliminate much of the con- fusion which must necessarily result should the term be accepted for any single disease, and the great indefiniteness which has clustered around this term as a name for a single disease will be cleared away. By the application of appropriate names to carefully discriminated con- ditions of the plant, much progress will be made in the understanding and treatment of these troubles. The purpose of the present article is to present a resume of the results of the investigations upon cotton diseases in the United States, some of which have already been published. These diseases maybe classed iu three general divisions, according to their etiology. Diseases due to physiological causes. — Mosaic disease, or yellow leaf blight, red leaf blight, shedding of bolls, and angular leaf spot. Fungus diseases. — Frenching; sore shin, damping off, or seedling rot; anthracnose; root rot; cotton-leaf blight; areolate mildew ; cotton-boll rot; and ripe decay of bolls. Nematode diseases. — Root galls. MOSAIC DISEASE, OR YELLOW LEAF BLIGHT. The later stages of this disease probably form the larger part of the troubles which are termed "black rust." The name mosaic disease, or yellow leaf blight, is quite characteristic of the early stages of the trouble as it is here defined, and renders it possible to differentiate it 279 280 THE COTTON PLANT. readily from the other troubles, which are often spoken of as " black rust," but which are in reality quite different in their nature. The term "yel- low leaf blight" was first used by the author in 1S92.1 "Mosaic dis- ease" was added to tliis term, or used synonymously, a few months later.2 The latter seems the more appropriate, but since the former was first used in differentiating this peculiar disease from the others it seems well at least to continue its use in the literature of the sub- ject for the present. During very rapid progress of the disease also the mosaic character of the leaf is not so apparent as during the normal development. In 1891 a preliminary investigation of the so-called black rust was made.3 The study was confined entirely to the organisms present on the leaf and other parts of the plant, and it was not possible at that time to do more than to record the presence of certain fungus organ- isms, to observe their botanical characters, and to note the fact that their presence at least hastened the destruction of the plant. The following year investigations taken up at the beginning of the season confirmed the view that the organisms hastened the destruction of the plant, and at the same time demonstrated the fact that the organ- isms did not initiate the disease but only aggravated it. The results of the trials of Bordeaux mixture, eau celeste, and cop- per sulphate indicated that this disease could not be prevented by the application of fungicides, and confirmed the conclusion, drawn from observations of a different character, that it was due to physiological causes. Experiments conducted under the direction of the author in several localities in Alabama during two seasons showed a considerable reduc- tion of the disease on plats where kainit was the fertilizer used. At Auburn an experiment was conducted on three plats. Plat No. 1, on which cowpeas had been grown, received before plowing a heavy dressing of kainit and acid phosphate. No nitrogenous fertilizer was applied. Plat No. 2 received nitrate of soda in addition to other fertil- izers but no kainit. Plat No. 3 received a complete fertilizer. In July there was a perceptible yellowing of the plants in plat 1, while plats 2 and 3 bore a rich green foliage. The yellow color of the plants in plat 1 was evenly distributed over the leaf, there being no indication of the mosaic arrangement so characteristic of the disease. In Septem- ber the plants were matured and only a few showed any sign of the disease. The yellow color of the plants was due to the acid phosphate and kainit ripening the plants prematurely (acid phosphate being known to produce this effect), along with a suffused yellowing of the plants. Early in August the plants in plats 2 and 3 were badly affected, the 1 Alabama College Sta. Bui. 36. -Alabama College Sta. Bill. 41. 3 Alabama College Sta. Bui. 27; Bot. Gaz.; 10 (1891), No. 3, pp. 61-65. DISEASES OF COTTON. 281 leaves showing the checkered appearance of the disease, and were an easy prey for such fungi as Macrosporium nigricantium and Gercospora gossypina, resulting in their curling up, drying, and falling off. In a field of cotton of 3 or 4 acres near the scene of the above experi- ment the plants in May and June were very promising, but in August the disease had appeared to such an extent that the yield fell off at least one half of what would have ordinarily been expected. The fertilizer used in this case Avas stable manure, cotton seed, and acid phosphate. These experiments seem to show what has for some time been held by a number of intelligent planters who have experimented with kainit as a fertilizer. It has been quite frequently noted that with quite large applications of kainit there was no appreciable increase in the yield of cotton. This occurs in those seasons when the rains are quite fre- quent, not long continued, and keep the soil moist and the plant in normal growth. On the other hand, during dry seasons as well as sea- sons of drought followed by long-continued rains, kainit has a percepti- ble, sometimes a remarkable, influence in increasing the yield. This, with the well-known effect of such salts in changing the physical con- dition of the soil, leads to the belief that the increased yield and the comparative freedom from disease result from the action of the kainit in binding more firmly together the soil particles, so that it is more retentive of moisture or more able to draw it up from below.1 Salt and wood ashes are known to produce much the same results in the soil.2 Boiling the land is frequently resorted to in order to produce the same effect. In the cultivation of cotton the more progressive planters are careful to prepare the land well before planting, and then to cultivate only the surface soil afterwards, in some cases scraping the surface of the soil with a "sweep" to a depth of only a few inches. This leaves the underlying soil undisturbed, and there is no break in the continuity of the surface film on the soil particles below the few inches which have been stirred. The few inches of soil which have been stirred thus act as a mulch. Characters of the disease. — In the normal and usual progress of the disease there first appears a peculiar yellowing of the leaf, which gives it a checkered or mosaic appearance. The yellow color appears in small areas and bears a definite relation to the venation of the leaf, being bounded by veinlets which subtend areas more or less rectangular in outline. The green color is found along the larger and intermediate veins. The portions of the mesophyll lying along the veins, being near the channels for the distribution of the nutriment, receive a better sup- ply of moisture and assimilative material than the areas farther away and those along the smaller and terminal ramification of the vascular channels at a time when the supply is being cut short because of unfa- vorable conditions of the soil. They are thus enabled to hold the green 'Alabama, College Sta. Bui. 36. 2 See article on climatology and soils, p. 160. 282 THE COTTON PLANT. iff? flO"- color and continue the activities of the leaf for a longer period, while the angular areas most remote from the sources of supply are the first to feel the loss, and the deficient nutrition is manifested by the yellow color of the parts. During the first stages of the disease this color may become very pro- nounced, but later it may be marred by the appearance of discolored spots produced by the growth of fungus organisms in the tissues, weak- ened by the failing nutrition of the plant. Soon, however, there appear minute brownish spots in the yellowish areas, which increase in sizecen- trifugally, assuming a cir- cular outline and marked by concentric rings. The concentric rings are prob- ably due to the periodic growth of the fungus threads within the tis- sues, the periodicity being produced by variations in the temperature. The first fungus, which in most cases appears following the mosaic condition of the leaf, is Macrosporium nigricanti um A t k . As the leaf thus becomes in a badly diseased condi- tion, the Macrosporium is likely to be soon followed by an Alternaria.1 The black hyphse and spores of these two fungi soon give a black appearance to nearly the entire leaf, from which the disease takes the name of "black rust." These are not, however, the only fungi which are found as accompaniments of the later stages of the disease. Colletotrichum gossypii Southworth is sometimes found, and Cercospora gossypina Cooke, as well as its per- fect stage, Sphcerella gossypina Atkinson, is a very common accom- paniment of the trouble. The accompaniment of the Cercospora stage of Sphcerella gossypina frequently produces a separate type of the disease, especially when this fungus is more abundant than 1Thia may be Alternaria tenuis Nees, which Gasparrini found with other molds as an accompaniment of the disease of cotton in Italy known as Pelagra. (See Gaspar- rini, Osservationi sopra una malattia del cotone, etc. Inst. D'Incoraggiameuto. Napoli, 1865.) X Fig. 4.— Mosaic disease, or yenow leaf blight — Macrosporium form. DISEASES OF COTTON. 283 either the Macrosporium or Alternaria. This usually occurs when the disease progresses quite rapidly through the earlier stages, so that the yellow color is soou diffused somewhat evenly over the entire leaf or a large part of it. The Colletotrichum gossypii Southworth will be described under the paragraph on anthracnose of cotton, and the Sphcerella gossypina under the paragraph on cotton-leaf blight. Brief descriptions of the others may be given here. Macrosporium nigricantium Atkinson. The technical description is as follows : 1 Hyphse amphigenous, subfasciculate, or scattered, 0.050-0.140 mm. by 0.006-0.007 mm.; nodulose, septate, olive brown. Conidia 0.018-0.022 mm. by 0.036-8.050 mm., strongly constricted about the middle, stoutly rostrate at one side of the apes, smooth, transversely, longitudinally, and obliquely septate, olive brown. From the intercellular mycelium in the leaf the hyphse emerge to the outside, and, projecting a short distance from the surface, bear the spores at their free ends. When a spore is being produced, the hypha becomes somewhat enlarged directly below it. When the spore falls away, the hypha elongates at the end, the new growth arising from the inside of the enlargement, its contour being the same size as that of the hypha just below the enlargement. When the new growth first appears, it seems not to be connected with the enlargement, but projecting through it. As the hypha ages this appearance usually is not present and the enlargement seems to taper above into the new growth. The new growth thus formed at the end of a hypha may bear another spore, and so in favorable weather from two to eight or more spores may be borne in succession from a single hypha, the points where the successive spores were borne being just at the upper end of the successive enlarge- ments. At the enlargements there is usually a darker band around the center. The hyphae thus have a nodulose appearance in such spe- cies as Macrosporium parasiticum Thiim. As the young spore develops it is constricted in the middle before the first transverse partition is formed. The species of Alternaria has not been determined. The hyphse are scattered or loosely fasciculate, and plain, not nodulose, and by this character alone may be differentiated from the Macrosporium. The conidia are obclavate and borne in chains. In making microscopic preparations from the fungus on the leaf it is difficult to find the conidia in this relation, since they so easily become separated, but by making water cultures on a glass slip this manner of development is readily demonstrated. Other saprogenous fungi frequently appear in the final stages of the disease, but there is no especial interest or importance in discussing them. 1 Bot. Gaz., 16 (1891), No. 3, pp. 61-65. 284 THE COTTON PLANT. If dry weather continues for a long period and the disease is devel- oped, it is not an infrequent occurrence for the processes of dissolution in the leaf to continue so that the leaves curl up and fall away, and the plant may even die without the appearance of any of the organisms above described, or only a small percentage of the growth which is sure to appear if warm rains follow a drought when the disease is present. RED LEAF BLIGHT. The foliage of cotton frequently presents a red coloration, which is of the same nature as that displayed in what are termed " autumn leaves." It is an exceedingly common occurrence toward the maturity of the cotton, even of quite healthy and rank growth. It is of rarer occur- rence, however, in alluvial and rich soils than on poor lands. It is especially common on what is known as the " upland," where the soil is worn and poor. Here it occurs quite early in the season, and cotton some- times makes but little progress before the leaves become red, growth ceases, an early maturity sets in, and the leaves drop, while the plant bears from one to two or several bolls. The affection, if it can be so called, is usually denominated "red rust." It results from an impov- erished condition of the soil, showing a lack especially of potash and nitrogen, and probably also of phosphoric acid. This can be remedied by proper fertilizing and cultivation. A species of mite, probably Tctranychus telarius or a closely re- lated species, frequently produces an injury to the leaves of cotton which causes them to redden. At the same time the under surface of the leaves, where the mites effect the direct injury, is made to appear "rusty" or "scurvy," and, with the coloration of the leaves, would naturally lead to the designation of this trouble also as " red rust." It occurs in several of the cotton -producing States, especially North and South Carolina, where it has been known to do considerable injury. At one time, before the careful tests with certain fertilizers were made and before a careful study of the physiological relations of the plant to certain physical and worn-out conditions of the soil, the author was inclined to believe that this mite was perhaps the sole cause of the so-called "red rust." That it does sometimes cause serious injury which should not in any way be confounded with the red leaf blight is certain. Some planters believe that this phase of the disease, though they do not distinguish it from the red leaf blight, is more apt to occur near barns or buildings where there are numerous weeds, while others believe that it is certain to occur where the cotton is planted adjacent to a clover field, and still others say that if an armful of clover be car- ried across a cotton field the disease will soon appear all along that tract. This would favor the possibility that the mites in these cases were on the clover, and it is known that in some cases they are quite common on this plant as well as others. DISEASES OF COTTON. 285 Literature. — "The cotton worm and other enemies of the cotton plant," G. F. Atkinson, South Carolina Dept. Agr. Monthly Kpt., Oct., 1888, p. 91. South Caro- lina Sta. Bui. 4 (n. ser.), p. 60. Alabama College Sta. Buls. 36 and 41. SHEDDING OF BOLLS.1 The shedding of bolls or "forms," or their death and drying while still attached to the plant, is very frequently a source of great loss to the cotton crop. The trouble has been long known, but one widely prevalent and disastrous form has been misunderstood. It is often confused with the work of the bollworm, with punctures made by some hemipterous insect, etc. That some of the shedding is due to the work of the bollworm is true, but the shedding referred to here is a purely physiological trouble. During three years' observation in Alabama the author found this physiological form of shedding to be very serious. It occurs most fre- quently in extremes of either dry or wet weather, or during the change from one extreme to another. It may occur to some extent under nor- mal climatic conditions, especially if the cotton plants are too thick, or the variety of cotton is one which develops a very large amount of fruit forms in proportion to the leaf surface. During a normal period of growth the plants put out as many fruit forms as could be matured should the conditions favorable to growth continue. If a very dry period succeeds this, interfering with the sup- ply of nutriment and moisture, there will occur a partial withholding of tissue-forming material and moisture at a very critical period in the life of the young " forms," and the tissues of the young fruit are forced into an unnaturally matured condition. The fruit, including the peduncle and often more or less of the surface tissue of the stem at its point of attachment, becomes first of a paler green color than the adjacent parts of the plant, so that a well-marked color line delimits the healthy from the unhealthy portion. In many cases the tissue is separated at this line, so that the fruit falls off completely or hangs by a few fibers to the stem. The early growing season may be exceptionally favorable for the development of a large plant with an abundance of young fruit, and if followed by even ordinarily normal conditions will result in a j)artial loss of this fruit. A long rainy season may also cause the young bolls or forms to fall, the soil being so saturated with water as to interfere with root absorption, and the assimilative activity of the leaves will also be disturbed. The more or less complete separation of the tissues at the line of division between the healthy and dying portion depends upon the point of attachment of the fruit to the stem, and also to some extent upon the variety. In some cases the line of separation is apt to be clean cut, resembling the scar left by a falling leaf, especially when the peduncle stands at a strong angle from the supporting branch near its junction 'Alabama College Sta. Bui. 41. 286 THE COTTON PLANT. with the stem or larger branch. If the point of attachment is at a somewhat greater distance from this junction and the peduncle much more inclined obliquely, the line of separation is apt to include from one-half to 1 inch of the surface tissue of the stem below the pedun- cle, and very frequently then the lower part of the dead surface tissue does not entirely separate and the boll usually remains clinging to the plant. In some varieties, especially the cluster varieties of cot- ton, the separation of the tissues does not take place so frequently, and the boll usually remains firmly fixed in position; but the dead part readily indicates the tissues involved. The matured bolls do not form a separative layer of tissue when they mature, but remain fixed to the plant. The falling away of the dead immature bolls and forms, when it does occur, is a useful provision of nature, since the plant is left in better condition for the gathering of fruit which does mature. One great objection held by some to the cluster varieties of cotton is their tendency to hold the dead immature fruit. There is need of careful observations on this disease in order to throw light upon its treatment. ANGULAR LEAF SPOT. This disease is named from the dark angular spots which appear in the leaf. It is very widespread, but rarely appears to such an extent as to attract attention. Careful observation would probably reveal it in every cotton field during the growing season from May to July, and frequently later. The disease is first manifested by a watery .appear- ance in definite areolate spots, which are bounded by the veinlets of the leaf. The spots are sometimes very numerous and frequently con fluent. Often the disease follows one or more of the main ribs of the leaf, being bounded on either side by an irregularly zigzag line. In time the spots become blackish and then brown, and are frequently bordered by a blackish color where the disease has extended centrifu- gally. The dead spots in the leaf sometimes break out, leaving many perforations with ragged edges, somewhat as often results in "cotton leaf blight." The disease hastens the falling of the leaves. In the very earliest appearance of the spots, when the watery condi- tion is coming on, these spots swarm with bacteria. This suggested that it might be a bacterial disease. Cultures of the organism present were obtained and inoculations of healthy leaves were made at different times, but without producing the disease. It usually appears only in the older leaves which have passed their prime. It is quite likely that the bacteria present may easily start the trouble in such leaves, while they may be unable to affect the younger and healthy leaves. This may account for the failure of the inoculations. Sometimes, but rarely, it attacks all of the leaves on a plant. This suggests that such plants may be constitutionally weak from some unfavorable condition which renders them susceptible of attack. The trouble has not, however, DISEASES OF COTTON. 287 been sufficiently studied. There sometimes occur on the same plants bolls which present spots of the same watery appearance, which finally terminates in a rot and death. In this case the physiological weak- ness of the plant would naturally extend to the boll. A careful bacteriological study of the trouble in connection with physiological studies of the plant as related to different conditions of the soil, ferti- lizers, and meteorological conditions might throw more light on the disease. Literature.— "Black rusfc of cotton," G. F. Atkinson, Bot. Gaz., 16 (1891), No. 3, p. 16; Alabama College Sta. Buls. 27 and 41. FRENCHING. This disease has been known for some time among planters in the middle and southern part of Alabama. Inquiries among planters by the author regarding the origin of this singular name for the disease did not elicit any information as to either the time when the name was applied or the meaning of ihe word in this connection. Judging from the significance of the verb of the same root, which means something foreign, and therefore in a later sense strange, unnatural, etc., it is intended to denote a strange or unnatural appearance of the cotton plant. This disease is probably distributed over a large portion of the South- ern cotton belt. The author knows of its occurrence in the State of Alabama at the following places: Mathews Station, Hope Hull, Allen- town, Pike Eoad, Athens, Selma, and Montgomery. In 1892 specimens were received from Pine Bluff, Ark. The disease is sometimes con- fused with what is more properly known as variegated cotton, the foliage of which presents quite large angular areas of various colors, as red, white, and various shades of green on the same leaf, reminding one of a coleus plant. The author knows of several planters who thus confuse the variegated cotton with the disease frenching. Pammel1 speaks of the term "frenching" being applied to variegated cotton in Texas, though he records nothing like this disease. It probably occurs in that State, however. The first sign of the disease is usually a light yellowing of the lower leaves at the edge, or more commonly between the forks of the main ribs of the leaf. This yellowing of the leaf, which is sometimes nearly white and quite pronounced, is the result of a failing nutrition of the leaf, but the trouble is more serious than in the mosaic disease and pro- gresses more rapidly. It begins at the edge of the leaves farthest from the large veins, and then progresses rapidly up the leaf between the ribs. The leaf then presents the yellow color in a radiate fashion par- allel with the larger veins, and not in checkers, as in the mosaic disease. Quite early in the disease the leaf begins to brown at the points where the yellow first appeared, so that a brown color of the dead portions of 1 Texas Sta. Bui. 4. 288 THE COTTON PLANT. the leaf follows quite closely behind the yellow. In this way there are three distinct colors, green, yellow, and brown, in parallel radiating bands. The brown and dead parts of the leaf soon break out, leaving the leaf quite ragged. The green color lies along the sides of the veins. This is bordered by the yellow, and the brown cuts a V-shaped figure in the yellow area, while the entire margin or only a part of it may also be brown and dead. While the lower leaves are usually the first ones to be attacked, they do not go through all these changes of color before others are affected; but the general progress is from the lower leaves to those higher up on the plant. It is quite a common thing to see quite large plants with nearly all the leaves affected, the lower ones nearly dead while the upper ones are in the first stages of the disease. When the leaves are nearly dead, the tissues of the petiole at the junction of the branch mature, form a separative layer, and fall away. This may continue until all of the leaves of a plant fall off. The author's first observations were made on plants about 1 foot in height, a short period before the first blooms appeared, in 1891. In May, 1892, at Mathews Station, he observed the disease in very young plants, only a few days old, and before the plumule was developed. The peculiar yellow color was well developed in the cotyledons. These plants were growing in the "gunpowder" soil of the black belt, and the season being at that time a little dry, the soil would frequently adhere in a hard lump about the roots of the plants. Taking advantage of this, a few of these very young plants which showed the disease in the cotyledons were taken up and transported to Auburn, Ala., a distance by rail of nearly 100 miles, where they were transplanted. The removal checked their growth for a few days, then growth set in and all external signs of the disease disappeared. But in the latter part of June, when the plants were almost 1 foot in height, they were severely attacked, and by the middle of July all of the leaves pre- sented the striped appearance so peculiar and characteristic of the dis- ease. The disease has never been known, except in this case, in the region about Auburn. When the plant is old, these progressive changes in the color of the leaf are frequently distributed over many more courses on the leaf, following not only between the main four or five veins, but also the spaces between the primary branching of these veins. On plants pos- sessing a mild type of the disease, some of the leaves may exhibit the yellow color in indefinite courses — now a few large yellow spots some distance from the edge, or a pale yellow occupying nearly one side of the leaf. But the disease is always sufficiently characterized — either by the usual relation of the different colors, or by the peculiar shade of yellow, or by both — for one who has once carefully observed the disease to easily detect it, except in some cases described later. The sure test of the disease, however, is found by breaking or cutting DISEASES OP COTTON. 289 the stem of the plant. If it is "frenching" the tissues of the fibro- vascular system will appear light brown in color, the intensity of the color depending upon the virulence or stage of the attack. Planters say the heart is black. A microscopic examination shows the presence of a fungus, the threads of which sometimes completely fill some of the vascular ducts of the plant. The discoloration of the tissues is more pronounced in those ducts in which the fungus is located. Under the microscope the color of the tissues appears to be a brilliant yellow, unless quite old, when it is much darker and suffused with brown. The threads of the fungus are colorless when young, but become a bright yellow in age, and measure 2 to 4 /u in diameter. Minute spores, measuring 1 to 2 by 2 to 4 //, are developed from the ends of the threads, and are found either attached to their points of origin or free within the ducts. Pure cultures were obtained and in all such cases the fungus proved to be a species of Fusarium. The fungus enters the plant near the surface of the ground, or in the upper portions of the roots. As the threads increase they grow upward, and, reaching the branches and petioles of the leaves, grow out into their circulatory channels. This explains why the lower leaves are the first to be affected during the early period of the disease. The plants sometimes put out new growth after losing all their leaves, and seem to a certain extent to recover from the disease. In many cases the upper part of the plant dies and the new growth comes from the latent buds and dwarfed branches near the ground. These periods of convalescence may be followed by relapses several times during the season. The second attack often differs greatly from the first in external appearance, probably because the fungus is so well distributed all through the plant that its effect in attacking the new growth and increasing in the old is more rapid, thus not permitting the gradual sequence of color which takes place as described above. A few leaves sometimes show the characteristic sequence of color, but the leaf wilts, thus checking the color changes. The fruit is also affected and frequently undergoes decay when nearly ready to open, even on plants which do not seem to be very badly affected, or it frequently happens that in a light attack much of the fruit comes to maturity and opens in the normal way. Unless complicated with some other disease the fungus does not advance with such rapidity into the roots. This probably explains why so many plants sometimes recover, the roots in favorable weather supplying constantly the necessary moisture and nutrition and furnish- ing material for the growth of the latent branches near the base. In sandy land the disease seems to progress much more rapidly, especially when the plant has attained considerable size. It then often happens that very few of the leaves show the gradual color changes described above, but on a hot or dry day suddenly wilt, a few of the leaves wilting on one day and more on the following, or sometimes all on the same day, the plant then soon dying. 1993— No. 33 19 290 THE COTTON PLANT. This peculiarity of the disease by which the leaves suddenly wilt, in sandy land, is in external appearance very much like the root rot of cotton in Texas, but the etiology of the disease is very different. Occasionally the plants on sandy land are also affected with the root gall nematode. Whenever cotton in frenching districts is infested by these worms, almost every plant is also affected by the organism of frenching. This is probably because the roots, being diseased by the worm, offer easy access to the fungus. But many of the plants that are frenching are not affected by the worms, even in sandy land. The two diseases are quite distinct, but when both attack a plant the trouble is far more serious. " Knotty swellings," such as are caused by this nema- tode, are reported by some planters to occur on the roots of cotton in the soil of the black belt. These are probably caused by the same nematode, but the author has never observed them on the cotton roots in the prairie soil, while he has many times seen them on cotton roots in sandy soil. That they do occur in the prairie soil on the roots of other plants is certain, since they have been found on the roots of tomatoes and lettuce. Artificial cultures of the fungus. — Besides cultures for separation, which showed the fungus to be a species of Fusarium, cell cultures were made by using very thin sections of the diseased tissue, so that the microscope showed that the growth obtained originated from the fungus threads within the stem. In artificial cultures spore formation takes place within fifteen or twenty hours from the time of starting the culture. The hyphas in arti- ficial cultures usually remained hyaline, only one case to the contrary being observed. This occurred in the case of some hypha} in a bouillon culture which were left above the medium when transplanting some of the fungus from the bouillon culture. In a few days they had become the same color as the hyphse found within the tissues. In artificial cul- tures frequently enlarged cells occur as intercalary cells of the hyphse, which resemble gernmsB. Sometimes they occur at the end of a hypha, and in either case they may bear several flask-shaped basidia. In artificial cultures the spores measure 2 to 5 by 4 to 40 //, are con- tinuous or 1 to 5 septate, colorless, faintly granular, frequently possess- ing one or more vacuoles. The very minute spores are narrowly oval, and as they increase in length become curved. The shorter ones usu- ally have one end rounded, the other sharply pointed. The longer ones are usually pointed at both ends. The fertile hyphse, or basidia, as they are sometimes called, also vary greatly. The early ones, short and narrowly flask shaped, are sup- ported on the main hypha by a narrow pedicel. Later they frequently increase in size and branch profusely. The formation of spores in arti- ficial cultures reminds one of some species ol Glceosporiuni and Col- letotrichum where they are clustered about the end of the basidium. Frequently in this Fusarium, as in other species of this genus, the DISEASES OF COTTON. 291 fruiting hypha elongates as the spores are being borne and thus leaves the spores distributed along its course. On sterilized cotton bolls or Irish potatoes longer spores were developed than on agar or in bouillon. Parallel cultures were made of a saprophytic Fusarium which occurs on cotton throughout the cotton belt and is also frequently found on the bolls of frenehing cotton. The two seem to be specifically distinct. In the saprophytic species the spores are more strongly curved and the ends very long and slender. This distinction was maintained through- out several cultures. The name Fusarium vasinfectum was proposed for this fungus by the author.1 Inoculations. — Experiments were made in August, 1892, to determine if the disease could be obtained by inoculations with pure cultures Of the fungus. The Fusarium was considered not to be a sufficiently aggressive parasite to be able to make its way into the ducts of the circulatory system unaided. The fact that the "sore shin" fungus could disease the stems of young cotton plants, and that many of the plants recovered even after the ulcer reached the vascular system, sug- gested that possibly this fungus might prepare the way for the entrance of the Fusarium. Several tests were made on growing plants that had been inoculated with the sore shin disease. In one experiment pure cultures of the Fusarium were inserted with success, one plant showing the external characteristics of the disease, and the fungus was found in the tissues when a microscopical examination was made. In another experiment plants affected with the frenehing disease were placed in contact with some that were suffering from sore shin, the leaves of which in one case passed through the yellow color changes and then wilted. The ducts of the stem also presented all the characteristics of the disease. The result was more satisfactory than that obtained from the inoculations from the pure culture of the Fusarium. This suggested the possibility that bacteria which are frequently found in the diseased tissues were the cause of the disease rather than the Fusarium. While positive assertions in favor of the Fusarium being the cause of the dis- ease instead of bacteria can not properly be made, the evidence thus far in hand gives greater support to the former view. The Fusarium is invariably found in both cotton and okra affected with the disease. Bacteria are not always present except in later stages, for in quite a number of transplantings of diseased tissue to nutrient agar no bac- teria were developed, while the Fusarium always appeared. Again, the same species of bacteria did not always appear, but sometimes one and then another. The season of 1892 being the last one of the author's stay in the South, it was impossible to pursue the investigation further The separation of the bacteria and tests with each of the species pres- ent in the study of the etiology of the disease might be undertaken with advantage to clear up this point. 'Alabama College Sta. Bui. 41. 292 THE COTTON PLANT. Literature. — "Frenching of cotton," G. F. Atkinson, Agr. Jour., Montgomery, Ala., Aug., 1891. "Additional note on frenching of cotton," ibid., Sept., 1891. ''A season's observation on frenching of cotton," ibid., Oct., 1891. Alabama College Sta. Bull. 41. SORE SHIN; DAMPING OFF; SEEDLING ROT. These are names applied to a very common disease wliick causes young plants to rot off partially or entirely at or near the surface of the ground. There seem to he several phases of the disease. Some- times the tissues undergo a soft rot, which progresses very rapidly, and the early stages are not marked by any striking color characteris- tics. Another phase may progress rapidly or slowly and is usually quite well characterized by a reddish brown color which accompanies it. This phase is also characteristic in that it is usually manifested on one side of the stem in the form of an ulcer which gradually deepens until the vascular system is reached, when the life of the plant becomes really endangered. Even when this stage is reached, however, the plant may and does frequently recover. „ This latter phase is characteristic of a very common disease of seed- ling cotton. It is called by the planters in many places " sore shin." Many planters say that " sore shin " is the result of a mechanical injury to the plant from a cut by the " scrape " used in cultivation. The term is sometimes applied to such injuries upon quite large stalks of cotton, but it should not be confused with the " sore shin" of seedlings which is caused by the parasitism of a fungus. The fungus which is almost universally said to be responsible for the phenomena of " damping off" is Pythium debaryanum. While all cases of damping off are not by any means due to this Pythium, it is quite likely that many of the cases, of what above is termed a soft rot of seedlings, are due to it. The fungus with which we are chiefly concerned here will be called " sore shin " fungus for convenience, for it is not well known at present what the species of fungus is, or even the genus, for from all of the arti- ficial cultures yet obtained of it nothing but the mycelium and sclero- tium stage has been obtained. The diseased portion of the plant is just beneath the surface of the ground and presents an area of shrunken tissue of a dull brown or reddish color. The size of the shrunken area and the depth of the injury are proportionate to the serious condition of the ulcer, as it may be termed. If the injury remains confined to the superficial tissues the plant will usually recover. It does sometimes recover when the injury reaches the vascular tissue, but more frequently death results when the trouble has progressed thus far. When the study of the trouble was first undertaken an examination of the diseased tissue showed the presence of several fungi. Besides the frequent occurrence of Bhizopus nigricans and saprophytic species of Fusarium, there were generally present in great numbers nonfruiting DISEASES OF COTTON. 293 threads of some fungus. This led to the supposition of their causal relation to the disease. The threads are 9 to 11 // in diameter and the cells 100 to 200 // in length. At first they are colorless and pos- sess numerous vacuoles of varying sizes in the nearly homogenous protoplasm. Later they become brown in color. The branches extend obliquely from the parent thread, are somewhat narrower at their point of origin, and possess a septum usually 15 to 20 /« from the parent thread, giving a clavate form to this part of the branch which is con- tinuous with the parent thread. Frequently the hyphse are associated in strands, being woven and twisted together. Pure cultures. — By placing affected seedlings on filter paper in a moist chamber there are developed in twenty-four to forty-eight hours numer- ous threads in a horizontal or procumbent position, which extend out for 1 to 3 centimeters over the paper, often not contaminated with other fungi. By transplanting a few of these threads, using a flamed platinum needle, into nutrient agar rendered acid by lactic acid (1 drop concen- trated lactic acid to about 10 c. c. of nutrient agar), a pure culture of the fungus was obtained. A series of experiments was conducted to determine whether this fungus could really produce the disease and damp off the young plants. The experiments showed that the fungus used in the inoculations was the cause of the disease produced at that season in the gardens and fields examined. Numerous cultures were made on Irish potatoes, cot- ton stalks, oak wood, cotton seed, and horse dung, the details of the cultures, as well as the experiments mentioned above, being published in Alabama Station Bulletin 41. ANTHRACNOSE. (Colletotrichum gossypii South worth.) The fungus causing anthracnose of cotton was described in 1890 by Miss Southworth,1 according to whom the fungus was distributed in Ellis's North American Fungi,2 itnder the name Gleeosporium carpigenum Cooke. Type specimens of G. carpigenum were examined by her and found to be distinct from the cotton fungus. The presence of dark setre among the basidia separates it from the genus Glceosporium. During the same summer the fungus was studied quite independently by the author, who only learned of Miss South worth's study a short time before her publication appeared. His observations agreed with hers in placing it in the genus Colletotrichum. The result of the author's flrst study was read before the Association of American Agricultural Colleges and Experiment Stations at Champaign, 111., November, 1890, and later pub- lished in the Journal of Mycology.3 The fungus is probably very widely distributed, but serious injury 1 Jour. Mycol., Vol. VI, No. 3, 1890-91, p. 100. 2 Ellis's North American Fungi, No. 2267, from Louisiana. 3 Jour. Mycol., Vol. VI, No. 4, 1890-91, p. 173. 294 THE COTTON PLANT. seems to be confined to certain localities. The author has observed it at quite a number of places in Alabama, but only at Brundidge was any very serious injury to the fruit noted. At that place, in September, 1891, 10 to 50 per cent of the crop was destroyed on some plantations. In the vicinity of Auburn, while it occurs on other parts of the plant as well as the bolls, its greatest injury seems to be confined to the young plants. Character* of the disease on the hotls. — The disease on the bolls origi- nates in minute spots. These spots, when very small, are of a dull reddish color, and present minute shallow depressions of the surface tissue. As these spots eidarge the tissue black- ens until the development of the spores begins. These are developed in pustules, usually conflu- ent, in the center of the nearly circular spot. With their development the color of the spot changes. If there are a few spores it becomes a dirty gray, or a bright pink if the spores are numerous. Where the spores are few in number, many of them stand out upon the sur- face on threads which have grown up through the tissue. The spores being colorless, a grayish cast is given to the dark background of diseased tissue. When the spores are developed in great numbers they are piled up into a considerable heap and form a large confluent mass occupying the central portion of the spot. A pink pigment, given off by the spores, is produced here in such quantities that it can be seen. This gives the pink color to the spots. As the disease progresses the spots increase in size and the color bands which surround the spots move outward centrifugally. The outer band, which is the border of the spot, is dull reddish brown in color, and its outer limits are ill defined. Inside of this border is a blackish band, which borders the pink center. As the spots increase in size they frequently coalesce and form large, irregular, diseased areas, covering sometimes one-half the surface of the boll. If the fungus Fig. 5.—. Anthracnose. DISEASES OF COTTON. 295 attacks the bolls before they are full grown it arrests the growth of the tissues involved on that side of the boll, so that they are frequently inequilateral. It also induces a premature ripening- of the tissues, when the bolls become dead, and dry in fixed forms. The natural separation of the carpels is prevented. The boll either remains closed, or, as is more frequently the case, the carpels separate at the apex only, so that the boll remains partially open. In either case the fungus penetrates to the lint in many cases, and is often found upon it in great abundance. In such cases the seed is likely to be included in the attack. Several saprophytic fungi are accompaniinentsof the disease in these final stages. Affecting the stem. — So far as the author has observed the fungus does not produce any characteristic injury to the stem of well-developed plants which is noticeable, but it is frequently found in injured parts of the stem, and on the scars formed by falling leaves, where the dead tissue of the scar, especially in humid weather, invites its development. The fungus sometimes seriously affects the stems of seedling cotton, attacking tlie stem at the surface of the ground or just below, and caus- ing the plant to wither and die, much as if it damped off. The tissue reddens and shrinks frequently in longitudinal lines. The macroscopic appearances of the injury are usually quite different from those occa- sioned by the "sore shin" fungus. The stem is not ax?t to present the well-defined ulcer, or diseased depression, which is so characteristic of the injury from the latter. Seedlings are probably frequently diseased in this way from the spores which are lodged in the lint of the seed at the time of planting. In cultures of young plants in sterilized soil annoy- ance was sometimes caused by the development of the fungus under circumstances such that they could have been diseased in no other way than from spores which remained attached to the seed. Several times during the winter of 1892 and 1893 cotton seed from Alabama was planted in the forcing houses and botanical conservatory of Cornell University, and the fungus appeared sufficiently to damp off and disease several seedlings. This seed, which was gathered in the autumn of 1892, afforded a good illustration of the vitality of the fungus. Some of these same seed were planted during the winter of 1893-94 and the fungus appeared upon the stems of the young seedling. In all cases where the seed were scalded before planting the fungus did not appear. The anthracnose spores were not found in the lint in these experiments, and it may be some as yet unknown reproductive body accompanying the seed which will retain its vitality for such a long time. The an- thracnose spores, however, have been found to germinate when taken from the diseased bolls after five months. In trials of some from the same bolls at seven months the spores failed to grow. It is quite possi- ble that the mycelium may rest in the tissues of the seed, as in the case of the bean anthracnose, Collctotrichum lindemuthianum, and probably scalding the seed would not kill the mycelium within the tissues with- out also killing the seed, although this treatment might partially pre- vent the disease. 296 THE COTTON PLANT. Affecting the leaves. — The anthraciio.se is frequently found upon the leaves, it being more liable to develop in sickly leaves or injured places than to attack healthy ones. From the partial saprophytic habit of the fungus, otherwise diseased or injured leaves as well as stems provide a nidus for the propagation and transport of the fungus from the seedlings through the growing season to the bolls. The seed leaves, or cotyledons, however, suffer from a characteristic injury. While the seed is germinating, the spores caught in the tangle of lint still adhering to the seed coats germinate and attack the fleshy cotyledons as they are slipping from the coats. The fungus attacks the edges of the cotyledons and destroys an irregular area bordering the middle portion. The cotyledons, being quite fleshy and succulent, form a suitable place for the development of spores, and the diseased area is marked by the bright pink or roseate tint so characteristic of its profuse development on the fruit. The degree of success which attends the throwing off of the seed coats by the cotyledons during germination probably bears a very close relation to their susceptibility to disease. After the young root has emerged from the seed coat, or "hull," if the conditions are such as to cause the hull to dry and remain so, it is cast off by the cotyledons with difficulty and sometimes not at all. Frequently the hull clings to the extremities of the cotyledons, holding them firmly, while their bases are exposed to the light and consequently take on a green, healthy color. The edges of the cotyledons thus held acquire a sickly yellow color, and frequently the effort to extricate themselves results in. some abrasion of the tissue. In either case the edges of the coty- ledons, under such unnatural conditions, are an easy prey to the anthracnose spores which fall on them from the tangle of the lint still on the seed coat. Such cotyledons are sometimes attacked by a Fusarium, the spores of which also produce a pink pigment, and the fungus can then only be differentiated from the anthracnose by the use of the microscope. Young plants have been inoculated by sowing the spores on the cotyledons. This at first suggested the possibility of the fungus using this mode of entrance to the plant as a means of spreading through the tissues, to be prepared for the final attack on the fruit and other parts of the plant, much as is known to be the case in Cystopus Can- didas and certain of the Ustilaginea\ Examination of the stem of affected plants fails to disclose the mycelium in all parts of the plant, and there is no evidence that the anthracnose travels along through the plant from the young stem or cotyledons to the bolls and leaves. Circumstantial evidence is very strong, even, that this is not the case. At all stages of the growing season the fungus produces spores very soon after germination takes place, so that crops of spores are devel- oped in rapid succession where conditions for growth are present. Furthermore, the fungus is not in any appreciable degree an obligate DISEASES OE COTTON. 297 parasite, bat is markedly saprophytic at times. There is a reasonable possibility tbat the crop of spores produced on the diseased young- stems or cotyledons will soon find an opportunity for growth and pro- duction of another crop of spores at some injured point on a leaf, or upon the partly dying- tissue of leaf scars. When the fungus obtains a good hold in the tissue of the stem it does serious injury, which is not the case with tbe smuts in the stems of tbe cereals. Characters of the fungus. — The spores are oblong, sometimes rounded at both ends, but usually rather sharply pointed at the base. There is usually a broad, shallow constriction at the middle, so that the ends of the spore are greater in diameter than the middle. The protoplasm is distinctly granular, and one or more vacuoles are present. The spores measure 4.5 to 9 by 15 to 20 //. Where they are produced on green or decaying bolls or other succulent parts of the xdant they are usually associated iu distinct acervuli or heaps, which are 100 to 150 /.i in diameter. The acervuli are composed of numerous spores, with the fer- tile hyphae from which they are developed and tbe underlying stroma within the tissues of the plant. The fertile hyphae are of two kinds. Tbe most numerous ones are short, colorless threads, arising from tbe stroma, and standing parallel, and very closely crowded together. They are usually called basidia. The second kind are long, dark, olive-colored, septate setae, which are characteristic of the genus Colletotrichum. In tbis species tbey bear spores, wbich is quite an unusual tiring in the genus. The seta? are straight, curved, or flexous, simple or rarely branched. They measure 100 to 150 ju in length. Their distal ends are nearly hyaline, and the spores borne upon them are often obovate, tbe base being rather sharp pointed. Tbey seem to arise later in the development of the pustule after the ordinary basidia are developed, when parts of the stroma are becoming dark colored. They arise from the dark parts of tbe stroma or from rudimentary sclerotia. Many times in young acervuli the seta? are not developed, and if this condi- tion alone were known the fungus would be referred to the genus Gloeosporium. Artificial cultures. — Several artificial cultures were made to trace the development of tbe fungus. In some cases tbe nutrient medium used was agar peptone broth with and without an infusion of cotton leaves. Tbe spores germinate quite freely under favorable circumstances in from four to ten hours. At the time of germination or prior to it, fre- quently one or two transverse septa are observed in the spore, dividing it into two or tbree cells. Several germ tubes may be produced from a single spore. Tbe mycelial threads begin to branch immediately, and are somewhat flexous in their course. From all parts of tbe mycelium short fertile branches soon arise of one, two, or three cells length, wbich resemble the basidia and bear spores. Sometimes these fertile branches or basidia arise directly from the spore. In the solid medium the spores from a single basidium, when not crowded by the basidia and other 298 THE COTTON PLANT. spores, are clustered around the end, each succeeding spore pushing the one which lias just become free to one side. The sharply pointed basal end of the spore favors this. After several days there is a beau- tiful crown of spores clustered at the end of the basidium, all lying parallel to each other. Spores are sometimes produced within eighteen hours from the time of sowing. Besides the production of spores, certain branches, either remote from or near the center of growth, produce at their ends peculiar enlarged cells, olive brown in color and varying in outline, but always of greater diameter than the hyphae which bear them. These bodies frequently produce immediately a normal hypha resembling the others of the mycelium. This in turn may sooir produce another bud, or may grow to a considerable length and produce basidia and spores, or develop spores soon after its origirr from the bud like an ordinary basidium. In many cases the gemma immediately begins to bud in an irregular man- ner, producing cells similar in color, but very closely compacted together into an irregularly oval or elongated or flattened imperfect sclerotium. After one or two weeks' growth a large number of these gemma3 and imperfect sclerotia are developed near the center of growth. At the same time, the basidia have become very numerous at this point, arising from the mycelium or by the branching of older ones, and the mass of spores assumes a roseate tint. Cultures were also started in pure water and in weak nutrient medium. In water the germ tubes, when once or twice the length of the spore, almost invariably produced gernmae. If these developed other tubes it was only to give rise to other gemmae. In no case at that time were spores produced nor any appreciable length of mycelium. In the weak nutrient medium the gemmae were produced freely; also a number of hyphae produced spores. While the vegetative growth exceeded that of the spores sown in water, there was but little compared with that of spores sown in a rich medium, and the spores did not live so long. These gemma? are sometimes spoken of as secondary spores. They are not secondary spores in the usual accept- ance of that term. They do not become freed from the mycelium except by accident or by the dying of the thread to which they are attached, in which case they are more properly gemmae. Their frequent later development into compound gemmae by budding would sti engthen this view, and indicate that they are rudimentary sclerotia, or perhaps presage the development of pycnidial or ascigerous stages, as yet unknown in this genus. In cultures on nutrient agar the author has never observed setae to develop in such numbers nor so perfectly as they do naturally on the host. But by pouring a small quantity of agar on scorched lint in a culture tube and then inoculating this medium with spores the setae developed profusely. During October, 1893, cultures of the fungus were started again, this time from material from Mississippi, for the purpose of noting the form DISEASES OF COTTON. 299 and growth characters of the colonies in plate cultures. The dilutions were poured in Petrie dishes and kept near a north window, where it was rather cool, especially during- the night. The temperature ran up during the day to 70° or 75° F. The development of the colonies was rather regular in contour, though a marked periodicity of growth was induced by the change of temperature from night to day, resulting in the formation of concentric rings upon the colonies. The weft of fun- gus threads was rather thin over the entire colony, and radiating lines figured by the threads were plainly discernible even when the colonies were old. The first germ tubes did not branch profusely in the early growth, so that directly around the center of growth the colony was quite open, the threads not covering the entire space, and the light was transmitted through this part much more readily than through the portions of the colony surrounding it. The fungus was transferred to bean stems, and later to sterilized vetch stems. From this last culture another set of dilution cultures was started in April. They were kept at a warmer temperature, and while the growth of the colonies was similar to those grown at a lower temperature, the size of the colonies was greater and all the characters were shown on a greater scale. Some of the spores were quite small, so that only one or two germ tubes were developed at one end. Since the threads grew much more rapidly at the higher temperature and branching was not frequent or rapid at the early stage of development, the young colonies in these cases presented elongated, thin, and scarcely discernible tufts issuing from the point of growth. In other cases, where several germ tubes arose from all sides, numbers of these tufts extended in all directions from the center of growth. As growth progressed and the colonies increased in size, the periphery became more and more compact by the greater profusion in the branching. From the small spores where there were but few germ tubes, and these from one side of the spore, the colonies were un symmetrical, or one-sided, and the radiating threads formed well-marked, fan-shaped tufts. These fan-shaped tufts are also frequently developed in the symmetrical colonies where the original lines of growth from the spore are not sufficient in number to close up the periphery of the colony. This is more marked where the colonies are more numerous, as in the first and second dilutions, while usually in the third dilution, according to the method used, there were but few colonies, the nutrient medium being not so soon consumed, and there being more room for the advance of the colonies, they are more apt to close up at the periphery and form quite compact colonies. The un symmetrical or eccentric colonies in this case form at first several beautiful fan-shaped tufts, which soon unite and form a reniform colony, later to be closed at the sinus, though the colony would still remain un symmetric. Literature. — " Aiithracnose of cotton," South worth, Jour. Mycol., vol. 6 (1890-91), pp. 100-105. " Anthracnose of cotton." G. F. Atkinson, Jour. Mycol., 6 (1890-91), pp. 173-178. Alabama College Sta. Bui. 41. 300 THE COTTON PLANT. BOOT ROT OP COTTON (OZONIUM). A preliminary account of the root rot of cotton was published by Pammel in December, 1888.' A fuller account of the investigations appeared in the following November.2 The following account is an extract of the latter, supplemented by the results of the author's own investigations. The disease is a true rot caused by one of the higher fungi, but noth- ing as yet has been found except the rhizomorphic and a sclerotia like condition, so that the affinities of the fungus are still unknown. The fungus was published by Pammel provisionally as Ozonium auricomum Link, but it is quite likely that it is not identical with that species. Certainly it is not identical with several specimens which occur in some of the European exsiccata?. It will here be spoken of simply as "Ozonium." There is a general belief among planters that certain soils only har- bor the disease. The investigations showed, however, that nearly all classes of soil are more or less subject to it. Planters suffer most from the disease in the central black prairie region of Texas. Its northern boundary is the lied River as far east as Paris, Tex., extend- ing in a southwesterly direction to San Antonio and thence westward. The counties of Montague, Wise, Parker, and Hamilton are the west- ern boundary in the north. A white rotten limestone, often cropping out, underlies the entire region. The soil of these black waxy prairie lands is very retentive of moisture, which is a condition favorable to the development of the fungus. The moisture is especially abundant when the limestone comes near the surface of the soil. The cotton fre- quently dies from the disease on the limestone ridges and slopes when none is affected in other parts of the field. For the discussion of the various theories which have been advanced to explain the cause of the disease before it was shown to be due to thei>arasitism of a fungus, reference should be made to Pammel's work. Characters of the disease. — The first indication manifested by the cot- ton plants of the activity of the fungus is the sudden wilting of one or more plants. This is usually first noticed in the latter part of June and early in July, though the time varies with the locality. Planters sometimes associate the dying of cotton with the appearance of flowers and the bolls, but from the condition of fields early in July it seems that it makes its appearance much earlier in the season. R. D. Black- shear, of Navasota, Tex., has reported plants dead from the trouble as early as May G, and certainly quite young plants are affected with it. The fungus has been seen on plants only G inches high. The sudden wilting of a considerable number of plants does not occur, however, until near the middle of June or later. The wilting of a single plant here or there in the field might be overlooked or be attributed to some 1 Texas Sta. Bui. 4. * Texas Sta. Bui. 7. DISEASES OF COTTON. 301 other cause. This explains why from general observation the disease is thought to appear first at a later period. From a single stalk which dies in May or June the disease may spread so that areas of consid- erable extent will be affected by the close of the season. The dead patches have no definite boundaries, but extend in all directions through the field, the black streaks and patches formed by the dead plants occasionally containing a few green plants. In passing through the belt where the disease is prevalent a striking contrast is observed between the areas made black by the dead plants, everywhere so con- spicuous in the fields, and the interspersed green areas of apparently healthy plants. Tbe suddenness with which plants die is governed somewhat by the atmospheric and soil condi- tions, rianters frequent- ly say that dry weather checks the disease. Dur- ing the dry weather in August, 1888, few plants were dying. In the latter part of August the rains set in, and then during intervals of sunshine large numbers of plants wilted. In June and July, 1889, it was again noticed that more plants gave the ex- ternal evidence of disease after a rainy day which was followed by warm sun- shine than during several days of dry weather. Healthy plants are fre- quently found close to diseased ones even late in the season, but it does not follow that such green plants are not affected with the disease, as has been shown to be the case by examination. For example, eight plants were found growing in a cluster, two of which had wilted, and in each case tlie taproot was covered by the mycelium. In two of the green plants the taproot contained an abundance of the fungus, and the plants would probably have wilted in a very few days. In two other cases a small amount of the fungus was found on the roots, while only a single one at that time was apparently exempt. Fig. 6.— Root rot. 302 THE COTTON PLANT. Every plant which presents the pathological conditions above de- scribed possesses the fungus on the roots, if we add to the sudden wilting of the plant the characteristic shrunken areas of the roots. Under the article on trenching it will be seen that -sometimes this disease causes the plants to wilt suddenly under atmospheric conditions similar to those which are most favorable to the sudden wilting of the plants from the Ozonium. If the roots of plants just dead from the disease be examined, fre- quently there are small "wart-like" bodies on the surface, which very often occupy the lenticels of the root. These are probably the same as the sclerotia-like bodies, which are described farther on. Plants which have not yet succumbed to the disease, but which are affected, .show the presence of the fungus on the root, sometimes in considerable quan- tity, so that it appears as if the root, or a portion of it, were covered with a whitish mold. The white threads gradually assume a brown color, beginning in the older portions. The taproot is usually the first to be attacked, and the point which the fungus first invests is some- where near the surface of the ground. It does not begin at the tip or near the tip of the root, and in many cases the lower portion of the root is free from the fungus, at least in the early stages. In cross sections of the diseased roots the characteristic threads of the fungus are seen to extend into the medullary rays and into the vessels. The threads within the tissues of the root do not become brown in color, and there are not the peculiar branching setSB, but the identity of the threads with the fungus upon the surface can be determined by the direct con- tinuation of the superficial threads with the internal ones. The fungus derives its nourishment from the living substances of the root, and also in its physiological processes sets up certain fermentations which kill the affected portions, causing partial decomposition, which results in the shrinking of the tissues and the formation of quite exten- sive depressed areas. The borders of these depressions show at first a red discoloration, which ultimately becomes brown. Near the surface of the soil an enlargement frequently is formed in which elaborated materials are apparently stored during the progress of the disease. From these enlargements new roots are frequently developed as the lower roots are placed under contribution to the parasite. These help, in favorable weather, to prolong the life of the plant, but are usually not sufficiently developed to prevent the collapse of the plant when the older roots give way. When the roots become seriously injured, root absorption of material from the soil becomes so diminished that it is not equal to transpiration from the large leaf surface, and the plant wilts. In the affected areas the disease spreads from year to year in a cen- trifugal manner, the fungus making its way through the soil from plant to plant. The lint of the diseased cotton is injured, the fibers are wider, and the spirals are fewer and more uneven than in lint from healthy plants. DISEASES OF COTTON. 303 Some have supposed that the disease could be trausmitted through the seed, but experiments have shown that this is not the case. Morphology of the fungus. — The threads which are found upon the surface of the. roots are usually associated in strands which course over the surface of the root irregularly and branch quite frequently. The internal threads are sometimes associated in parallel layers, but are not formed into distinct strands like those upon the surface. The cells are hyaline, usually short, and of a greater diameter than the separate superficial threads or those upon the surface of the strands. The threads which compose the strands are usually of a greater diam- eter in the inside of the strand and smaller upon the surface. From the surface of the strands there are numerous free threads which stand out at various angles. Many of them are quite peculiar and character- istic of this fungus. They terminate in a long slender point, and fre- quently possess opposite or verticillate branches of the same character. No fruiting condition in the form of conidia has as yet been determined, nor any form analagous to the fruit body of any known fungus. The Ozonium auricomum Link, of Europe, is very different from the above fungus, though it is similar in habit in many cases and produces diseases of the roots of various plants. The space is too limited here for a discussion of the diseases due to this fungus, but mention should be made of the supposed complemeutal or fruiting form which has been reported in a number of cases in Europe. The complemental fruiting form is given in many cases by different authors as a very distinct fungus, which is evidence that very much weight can not be given to the specific determination of these root fungi by the rhizomorphic form only, unless there seems to be some very characteristic features, as in the case of the Texas Ozonium, which mark it very clearly. The European Ozonium is said by Schroeter1 to be the undeveloped condi- tion of Coprinus radians Penzig2, of G. intermedins Winter3 and Sac- cardo4, of Tramates odorata Holuby5, of Agaricus diliquescens. This would seem to show that the European Ozonium was at least the rhizo- morphic form of one of the Hymenomycetes. Pammel is inclined to think that the Texas Ozonium is the undeveloped form of some Pyre- nomycetous fungus, probably being influenced in forming this opinion by the discovery of the frequent association of a fungus in the roots of the diseased cotton and sweet potatoes which possesses blackish rotund bodies that resemble the perithecia of some Pyrenomycetes, and also by a doubtful artificial culture which he obtained by washing the tin eads of the Ozonium. The cultures obtained from this were not in any way like those of the fungus as it appears in its natural habitat, 1 Kryptogameu Flora von Scblesieu, B<1. Ill, 1st Hefte, Pilze, 1889, p. 519. 2Sui rapporti j^enetici tra Ozonium et Coprinus, Nuovo Giorn. Bot. Ital., XII, pp. 132-143. 3Die Pilze, in Kabenborst's Kryptogameu Flora von Deutscbland, I, p. 405. 4Saccardo, Sylloge Fung., VI, p. 345. 5Zur Kryptogameu Flora von Ns. Podbrad, Oest. Bot. Ztscbr., 1874, No. 10. 304 THE COTTON PLANT. but it was suggested that this might be due to the influence of the artificial medium. The pure cultures obtained by the author later show that this could not be the case. Pammel himself did not place much confidence in the results from his supposed cultures of the Ozonium. But this makes little difference from the standpoint of the treatment of the disease. Treatment. — The results of experiments at the Texas Station1 show that the disease can not be controlled by any application to the soil at present known. Rotation of crops seems to be the only method which will keep the fungus in check. There are, however, a large number of other plants upon which the fungus can grow readily. If plants which are susceptible to attacks from the Ozonium are grown year after year on the same ground the soil becomes, after awhile, so thoroughly infected with the fungus that the crop will not grow to any extent. Corn, sorghum, millet, wheat, oats, and other members of the grass family are suggested as desirable crops to grow in rotation with sus- ceptible plants. Some suggest a rotation which will bring cotton or a susceptible crop into cultivation .on infected soil not oftener than once in three or four years. Other plants subject to the disease — Pammel names the following plants as subject to attacks from the Ozonium, and from which the disease may be communicated to cotton grown in the same soil: All nursery stock except species of the genus Prunus, apple trees, Russian and paper mulberry, China berry, Japanese persimmon grafted on native persimmon, elm, basswood, and silver maple. Of these the China berry and paper mulberry trees suffer most severely. In August, 1888, the roots of a number of living China berry trees were observed to be covered with the fungus, and in the following year they were found to be dead, and the disease had spread to some young paper mulberry trees in the neighborhood. Old and dying trees frequently develop suckers. At Anna a num- ber of paper mulberry trees nearly dead produced hundreds of suckers from 6 inches to 2 feet high which covered an area of several rods. The suckers were wilting in large numbers and many were dead. Specimens of the pear from Burnet County had the fungus on the roots, but the trees had been dead for some time and it was thought that the fungus worked only as a saprophyte on these trees. When young apple trees are affected with the fungus the leaves suddenly wilt and turn black and in a short time the trees perish. In older trees death is more gradual. They have a sickly appearance several years prior to death, bear a heavy crop of fruit, and then die. The Ozonium disease of apple trees must not be confounded with the trouble brought about by the presence of aphides (Schizoneura lanigera) on the roots, which develop, as the result of injury, irregular knots on the roots. Some- times both the aphides and the Ozonium are found upon the same root. 'Texas Sta. Bui. 7. DISEASES OF COTTON. 305 Weeds affected by the fungus. — The common sida (Sida spinosa) is very commonly attacked by the Ozonium in infected fields. Even in fields well cared for by the planter this weed may be found in limited numbers. Frequently in such cases as well as where the weed is more abundant it is found to be attacked by the disease. This would be expected, at least since other plants are known to be subject to the attacks of the fungus, because the sida belongs to the same family as the cotton plant. The fact that this weed is subject to the attacks of the fungus and that it is a very widely distributed and common weed throughout the cotton belt is one difficulty in the way of success from the rotation of crops. The ragweed and cocklebur and some other Composite were found to be affected by the fungus. In all the cases observed the attack upon these plants seemed to follow some injury of a mechanical nature. There was no doubt that the fungus caused the death of the plants in question, but it was considered doubtful if they initiated the trouble. In orchards and nurseries all diseased plants should be dug up and burned. Care should be exercised in. the purchase of nursery stock to see that the roots are perfectly healthy. In the use of sweet potatoes for seed great care should also be taken, for in the use of affected pota- toes or the planting of affected nursery stock the fungus may be trans- planted to soils in which formerly the fungus was not present. . To the above-mentioned plants which are subject to the Ozonium should be added alfalfa.1 ARTIFICIAL CULTURES OF THE OZONIUM."2 During the summer of 1891 the author made several attempts to secure pure cultures of this fungus from material received from Texas. The affected roots were placed on sand in moist chambers, and the fungus strands grew in several instances from 4 to 6 inches out over the surface of the sand. Portions of these strands were transferred to nutrient media and failed to grow. In 1892 the work was again undertaken, the author being supplied once or twice a week with fresh material. Each root before shipment was carefully wrapped separately Avith moist paper, retaining a small portion of the earth with it, several of such roots making a single package. Before receiving the specimens the author prepared several moist chambers in the following manner: A layer of sand about one- half inch deep was placed in a glass vessel and covered with four thick- nesses of filter paper; the sand and paper were then moistened with distilled water and the cover placed in position. The moist chambers were sterilized in an oven at a temperature of 110° C. for an hour or two for two successive days. On the arrival of the roots they were carefully unwrapped, the earth removed, the roots washed with dis- tilled water and cut into pieces 5 cm. long, and placed horizontally on filter paper, four or five sections in each chamber. In two or three 1 Texas Sta. Bui. 22. 2Bot. Gaz. 18 (1893), No. 1, pp. 16-19. 1993— No. 33 20 306 THE COTTON PLANT. days the strands of Ozonium could be seen growing out over the filter paper for a distance of 4 to G cm. Sterilized glass slides were now placed at the advancing end of the strands, and upon these were placed small sections of cotton roots which had been previously boiled and then steamed for several hours to sterilize them thoroughly. Sections of such roots about 3 cm. long were placed upon the glass slide so as to come in contact with the fungus. At the same time similar sections of sterilized cotton roots were placed in test tubes in moist sand and partially immersed in distilled water. In twenty-four to forty-eight hours the Ozonium strands would take hold of the bait placed before them, so that the section of root could be transferred bodily to prepared culture tubes. The end containing the growth was placed in contact with the sterilized root already in the tube. Sterilized sweet potatoes were also used in test tubes as media for transplanting. The fungus grew rather slowly, but out of 75 baits which were promising 4 or 5 proved to be pure. From these, cultures were multiplied to the number of 50. Great difficulty was encountered to prevent the contamination by several species of fungi and bacteria. The Ozonium on artificial media, such as sterilized cotton roots or sweet potatoes, grows readily after once securing a firm hold, and possesses all the other characteristics observable in a natural condition upon the cotton roots. Being free from obstruction and hindrances which it encounters in nature, the growth is perhaps more compact, numerous strands uniting to form a brown weft, but the peculiar strands were present as well as the characteristic and branched seta?. Since the first transplantings the author has become more familiar with the habits of the fungus, and has found it an easy matter to cul- tivate it with certainty and in profusion. The Ozonium grew steadily over the sterilized cotton roots in a broad weft, the threads of which radiated from the point of infection. The line of advance is rather irregular, due to the unequal growth of the filaments. The weft is white for several days, when numerous branches arise on the free surface, giving it a woolly appearance. Later, promi- nent strands appear, lying in irregular parallels. About this time the color changes to a light yellowish brown, the characteristic color of the older phases of the fungus. After considerable growth has been made sclerotia-like bodies appear. They are from very small to 3 mm. in diameter, whitish and woolly, finally becoming of the same color as the filaments of the fungus. Internally the sclerotia are hyaline, and the cell walls are very thin. The outer portion is composed of threads similar to those of the strands. The development of the sclerotia was checked, and the experiment was begun anew in the winter of 1892-93. Having no cotton roots on which to grow the fungus, apple-tree root, sweet potatoes, and horse dung were substituted as culture media. The fungus grew quite well, but failed to produce any sclerotia. DISEASES OF COTTON. 307 The structure of the strands and threads conformed essentially with Pammel's description. Some of the threads which form the early stages of the strands are much larger than others, and are strongly con- stricted at the septa. Branches of various sizes put out from the parent thread, and in their growth closely invest it in a sinuous and irregular spiral. Other branches put out from the first in the manner just de- scribed, the new branches following the general direction of the strand. The branching of the strands depends to a certain extent upon the branch- ing of the primary threads. The formation of the strands resembles the formation of the cortical layer of some of the species of Batrachospermum or the covering of the axial filament of species of Lemanea. In artificial cultures, near the ends of some threads were observed swellings resembling oogonia. Sometimes the enlarged end of the hypha is brought in contact with these oogonia-like bodies, as though to fertilize them, but nothing resembling the fertilized oogonium was observed by the author. The bodies here described quickly disinte- grate. At first this phenomenon was thought to be preliminary to the formation of some sort of fruiting body, but it seems more likely that the protoplasm, developing certain acids having a strong affinity for water, imbibes large quantities of water, and the branches become abnormally swollen before their disintegration. Besides the Ozonium in several cases another fungus was present, which has been supposed by some to be another stage of the one under discussion. In the case of this one the mycelium is dark in color, and dark perithecia-like bodies have been noticed by Pammel and the author, and the author noticed in some cultures a peculiar form of budding; but while the fungus was carried through several cultures at different times, none of the supposed fruit bodies produced spores, so that the exact position of the fungus is mere speculation. That it is different from the Ozonium there is little doubt, since in artificial cultures of two complemental forms of the same fungus there should be some indi- cation in one of them of the affinity. There was, however, throughout the entire series of cultures of both fungi no form or appearance which indicated the least resemblance that could be taken as indicating an affinity between the two. During the cultures many saprophytic fungi came under the author's observation which were associated with the Ozonium on the roots in the soil. Some of these were obtained as pure cultures on the baits which were set for the Ozonium. Some of these were sent to A. P. Morgan, Preston, Hamilton County, Ohio, for determination. The fungi noted in this connection were as follows: (Edocejtltalum echinu- latum Thaxter, Sporotrichum chlorinum Link, Penicillium candidum Link, P. glaueum Link, P. duclauxi Delacr., Mucor inucedo Linn., RMzopus nigricans Ehr., Licea lindheimeri Berk'?, Verticillium rexianum Sacc, a species of Eurotiuin, several species of Fusarium which were very common, the fungus described in the paragraph on damping off, several nonfruiting forms, a pink yeast, and several species of bacteria. 308 THE COTTON PLANT. It would be very interesting, from a scientific standpoint, to know the perfect or fruiting form of this Ozonium on cotton roots. If a pure culture could be again obtained, and then grown on a large num- ber of sterilized cotton roots all massed together in some sod in a place where moisture conditions would be favorable, in time this form might be developed. The failure to obtain this form in the above trials is attributed to the fact that all the cultures were carried on on too small a scale, so that there was not a sufficient amount of roots in one place, and the moisture conditions were subject to too great variations, so that in the course of a few weeks the culture would become partially dry or completely so. The environment should be such that proper conditions of moisture could be sustained for a year or more. The author's study of the fungus thus far, however, leads to the belief that when this form is discovered it will prove to belong to some genus of the Hymenomycetes. COTTON-LEAF BLIGHT. (Splucrella gossypiva Atkinson.) The cotton leaf blight is a very common disease of the cotton plant, but it very rarely becomes serious. It usually attacks only the older leaves or those which have become weakened by physiological disturb- ances affecting the nutrition or assimilation of the leaves. The fungus was first known iu what is called the imperfect form, or conidial stage. This form of it was described by Cooke as Gercospora gossypina. The diseased areas on the leaf are in the form of rounded, irregular spots, which possess a dull reddish border surrounding a brownish or whitish central area. In the central area the fungus is located. The inter- cellular mycelium consists of irregular, branched, septate threads, which here and there near the surface develop a tuberculate stroma that gives rise to tufts of emergent hypha?, brownish in color and several times septate. Long, slender, terete, septate, hyaline conidia are borne at the ends of the hypha*. After one hypha has developed a conidium this becomes freed and the hypha grows out at one side of the end and bears another conidium. This process is usually repeated several times, so that each hypha bears several conidia, and the hypha presents a toothed or zigzag form at the fr nit- bearing end, each tooth or angle representing the place by a scar where a conidium had grown. These conidia germinate by the y>rod action of a germ tube at any or all of the usually numerous cells. These penetrate the tissue of fresh leaves and spread the disease. The spots first appear as minute red- dish dots, which increase in size centrifugally, and finally the center becomes brownish, leaving the red border which is characteristic of the spots. As the spots become old the central portion frequently becomes broken out, leaving the leaves with a perforated and ragged appearance. This fungus quite frequently attacks the leaves of cotton affected with the mosaic disease. In some phases of the mosaic disease it is the commonest fungus on the leaves, and as the leaves die they become DISEASES OF COTTON. 309 1 .^ curled in various ways, and during favorable weather, as during rains in the hot summer time, a great profusion of hyphse and conidia are developed. The brown hyphae give a blackish appearance to the leaf, while the colorless conidia form a whitish covering to portions of it. The conidia and hyplme, under these circumstances, are very long. On such leaves the perfect, or Splmerella, stage of the fungus is quite likely to be developed. These have been found in several parts of Alabama, and they undoubt- edly occur in all the A cotton-producing j,?^ States. This stage /;;' of the fungus (Sphce- rella gossypina) was first described iu a bulletin of the Tor- rey Botanical Club.1 The perithecia are ovate and nearly black, and partly immersed in the tis- sue of the leaf, the ostiolum and the apical portion pro- jecting through the epidermis. They measure GO to 70 /u. The asci vary from clavate to lanceolate or subcylindrical, measuring 40 to 50 by S to 10 fi. They are eight- spored, the spores being ellipti- cal or nearly fusoid, and slightly constricted at the single septum, which divides the spore into two unequal cells. Literature. — Cooke, Grevillea 12 (1883), p. 31. " Cotton-leaf blight," F. Lainson- Scribner, U. S. Dept. Agr. Rpt. 1887, p. 355. "Sphserella gossypina," etc., G. F. Atkinson, Torr. Bot. Club, Bui. 18 (1891), No. 10. Alabama College Sta. Buls. 27, 36, and 41. Bot. Gaz., 16 (1891), p. 62. AREOLATE MILDEW OF COTTON. Fig. 7. — Arfiolate mildew. (Ramularia areola Atkinson.) This mildew is confined to definite areolate portions of the leaf, the areas being limited by the veiulets. The clusters of short hyphse which project through the epidermis to the outside of the leaf and the num- bers of conidia borne upon them give a mildewed or frosted appearance 'No. 10, 1891. 310 THE COTTON PLANT. to the spots. The hyphse are colorless, usually simple, though sometimes branched, and divided into short cells. The conidia are short oblong, one to three septate, the ends being usually abruptly pointed. They are borne singly or in chains, and the successive places on the hyplne where they were attached show a toothed surface. The fungus has been found by the author several times in different parts of Alabama, and specimens have been received from Mississippi. It is not of very common appearance, and the indications are that it is not harmful, though it is sometimes produced in considerable profusion. It probably occurs in other States also. It was first discovered in the autumn of 1889.1 When viewed from above the spots present a light yellowish tint in the green leaf. Viewed by transmitted light tne spots are paler and permit the light to come through more freely than through the other parts of the leaf. The hyphae are usually more numerous on the under- side of the leaf and sometimes only upon this side, but they may be borne upon both sides even on the same spot. It is not likely to become of any serious importance. Literature. — "A new Karuularia on cotton," Bot. Gaz., 15 (1890), p. 166. Alabama College Sta. Bui. 41. COTTON-BOLL ROT. (Bacillus gossypinus Stedman.) This disease was first described by J. M. Stedman.2 It affects the bolls, seed, and lint. Affected bolls were first received from Baldwin County, Ala., in August, 1893. The bolls were in a rotten condi- tion and contained iusects which were determined as Epurcca (estiva, and Carpophilm mutilatus, beetles known in the Southern States, Mex- ico, Central and South America, and having the habit of feeding upon decaying and injured fruit of all kinds, and sometimes found sucking the sap from wounds in trees. They were found common in cotton bolls and in heaps of decaying cotton seed. They were thus regarded as not having any connection with the cause of the disease. Saprophytic fungi and in some cases the anthracnose of cotton were found, but iu bolls that were only slightly diseased no fungus was observed. Pure cultures of the bacteria inside the closed bolls were made by the plate method, and cultivations were started on gelatin and agar. In gelatin in four days the bacteria clouded the entire mass, giving it a greenish color. In agar the growth produced a milky cloud along the entire track of the needle path and over the surface of the agar as a more or less white, semitransparent glossy growth. Bolls on healthy cotton plants were inoculated with the bacteria from these pure cul- tures, and in all cases the disease appeared. Other bolls punctured with a sterile needle remained healthy except in one case, where the 'Bot. Gaz., 15 (1890),p. 166. -Alabama College Sta. Bui. 55. DISEASES OF COTTON. 311 decay produced was different from that characteristic of the disease. This rot is said to originate within the boll, and is not apparent until the contents of the boll are decayed, when the carpels show signs of tbe -disease in places. It first begins as a small, dark-brown area involving the young seed at the point near the peduncle. If it begins some time before the maturity of the boll, the entire boll will rot and not open ; but it may begin so late that only a few seed and a small portion of the lint are affected, while the carpels separate and the lint may be exposed and gathered. Stedman discusses the question of the probable means of entrance of the bacillus into the interior of the boll. The suggestion that the germs may in the ground in some unexplained way make their entrance in the root and travel up the stem to the boll is made, and also the question of entering the young ovary at flowering time, when the germs may be distributed to the ovaries by the agency of insects. The possibility of the germs being in the seed at the time of planting is also suggested, and plans for experiments to determine the manner of infection are proposed. The organisms would not induce pathological conditions when introduced to any other part of the plant, although in some cases they were able to live for some time. The disease is chiefly confined to the middle and top crop, first manifesting itself early in August. The organism is a short straight bacillus, truncate at the ends, with slightly rounded corners, 1.5 pi to 0.75 //. It is usually solitary, some- times in pairs, and. occasionally in chains of three to four. It stains readily with the usual aniline colors, is aerobic, nonliquefying, motile, and forms spores, though the latter are not described. The problem of the entrance of the organism to the bolls should be investigated, together with the relation of the plant to soil and climatic conditions in connection with the disease; also its possible connection with the disease called " trenching " should be taken into account. In this disease the organism, as we know, travels up the stem through the vascular ducts, and aside from the foliage characteristics, which are usually quite marked, many of the bolls are affected, and the rot here begins in the interior of the boll at the same point as in the cotton-boll rot. It should also be borne in mind that when the bolls are within a few weeks of maturity they are in one sense a form of nutrient medium for even saprophytic fungi and bacteria which are introduced into them, so that too much reliance should not be placed upon the results of the inoculation of the bolls through needle punctures. ROOT GALLS OF COTTON. (Heterodera radicicola (Greef.) Muell.) The roots of cotton are sometimes affected with a disease caused by a nematode worm, which, living in the tissues, causes abnormal growths termed u galls." This is also a very common affection of many other 312 THE COTTON PLANT. plants, and especially in gardens throughout the South and in green- houses in the North. The worm is, in fact, quite genet-ally distributed over the world. A related species, Jleterodera schachtii Schmidt, is also quite common in Europe, affecting especially the roots of sugar beets, and its life his- tory has been well worked out by Strubell.1 Jleterodera radicicola was made the subject of investigation by Mueller,- though he left some important points undetermined. In the autumn of 1889 a paper on the injuries produced by this worm was published by the United States Department of Agriculture/1 Almost immediately following this paper was o n e by the writer4 (1889) dealing with the life history and metamorphoses of this nematode and the injuries produced by it upon the plants in the vicinity of Auburn, Ala. In the course of this work some points in the life history of the worm were more fully elucidated. The females are of a dull white or yel- low color, with irreg- ularly oval bodies from 0.25 mm. to 0.5 mm. in diameter. Th ey are usually quite easily differen- tiated when mature from the tissue of the plant in which they lie when the gall is broken open. Usually the unaided eye can also distinguish the head end projecting as a minute point on one side, Fig. 8.— Root galls. 1 Untersuchungen tiber den Bau mid die Entwickelung des Riibenuematodeu., Het- erodera schachtii Schmidt (Bibliotkeca Zoologica. Original Abhandlungen aus dem Gesauimtgebiete der Zoologie, R. Leuckhart uud C. Chun. Heft 2), Cassel, 1888. 2Mittheilungea iiber unseren Ktdturpflanzen schiidliche, das Geschlecht Heterodera bildenden Wiirmer, Laudw. Jahrb., XIII, 1884. :i The root-knot disease of the peach, orange, and other plants in Florida, J. C. Neal. (U. S. Dept. Agr., Div. Ent. Bui. 20.) ••Nematode root galls, a preliminary report upon the life history and metamor- phoses of a root-gall nematode, etc. Ala. College Sta. Bui. 9. DISEASES OF COTTON. 313 giving to the object the appearance of a minute gourd or "crooked-neck squash," or of a minute inflated bladder. They can be well seen with the aid of a small hand glass. The cephalic portion of the body is cylindrical, the end being rounded; the mouth is terminal, and opens caudad into the oesophagus which leads to the intestine. Lying within the mouth is a spear, slender and long-pointed at the cephalic extrem- ity, and ending in a three-lobed knob at the caudal end. It is capable of extension, being moved by pairs of muscles attached directly to it. In the males the caudal end of the spear is supported by six lamella?, the ends of which form the cephalic end of the head and fit around the spear. The oesophagus begins at the base of the spear. The cephalic portion is a long, slender, tortuous channel, which appears as a dark line, reaching to near the swollen part of the cyst. The middle portion of the oesophagus is an ovoid or ellipsoidal transparent, muscular bulb, which has a fibrillate structure, the fibrillar radiating from the center. The caudal portion of the oesophagus connects with the intestine. In the gravid females this is difficult to see, since at this stage the mass of fat globules renders the body cavity too opaque. Were it not for a slight movement of the apparatus just described as the spear is thrust back and forth, or slight movements of the head, there would be nothing to suggest what we ordinarily consider a sign of life. Occasionally while the cyst is under microscopic examination the spear is thrust slowly out at the mouth and then drawn back. At the same time the cephalic part of the oesophagus connected with it is also moved. Sometimes the apparatus slides far enough so that the tortuous portion of the oesophagus is straightened and the bulb is moved a little forward and backward. Sometimes there appears also a slight lateral motion of the head, a sudden, "jerky" motion. This is probably from force of habit, for in the larval stage movement from place to place is accomplished by a constantly changing, tortuous motion of the body. Mueller (loc. cit.) speaks of an expansion and con- traction of the middle part of the oesophagus which he has observed. By this means nutriment from the plant is sucked into the oesophagus and thence passed to the intestine. Within the vesicular portion of the body, frequently more or less obscured by the mass of fat globules, are the two long genital tubes, which are coiled and considerably longer than the body. The genital tubes are paired, and near the caudal portion of the body they unite into a common tube, which extends to the exit, the vulva. The genital tubes, though nearly cylin- drical, are composed of four parts. The anterior or free ends, including about one-third the length, are the ovaries. The receptaculum seminis is near the center, and is connected with the ovaries by the oviduct and with the vulva by the uterus. The mature male is nearly cylindrical, 1 mm. to 1.5 mm. in length and about 43 ).i broad near the middle. The body is a little less in diameter at the caudal end, and the cephalic half tapers very slightly to the head end, which is about half the diameter of the middle of the body. 314 THE COTTON PLANT. The body is beautifully marked by prominent transverse striae, broader and much more distinct than in the larval stage. The excretory canal opens on the ventral side a little caudad to the muscular bulb. The caudal end of the body is curved, and near it are the two spicules at the opening of the cloaca. The generative organ is paired. The long slender testes, lying on either side of the intestine, reach by their free cephalic ends to about the middle portion of the body. Some little distance from the caudal end of the body they unite into a common canal, which itself, near the spicules, unites with the intestine to form the cloaca. The spermatozoa are spherical. The cellular structure of the testes resembles to a certain extent that of the ovaries. The cells are polyhedral from mutual pressure. In living males the spherical sperma- tozoa are easily seen at or near the common passage, but they are developed in the caudal ends of the testes. By boiling infested potatoes so that the worms could be removed easily from the softened tissues without cutting or crushing them it was found that it toughened the tissues of the animals and made the cellular structure of the testes, by coagulation of the albuminous substances, very distinct. DEVELOPMENT AND METAMORPHOSES. Eggs. — The young ova are developed in great numbers in the ovaries, and when full grown the genital tubes are crowded for nearly their entire length. They are very plastic; when free are spherical, but crowded in the ovaries are polygonal. Each one presents a large nucleus and distinct nucleolus. When quite young they are nearly hyaline and transparent. Near the cephalic end of the ovaries there are several layers of them because they are quite small, but as they grow in size and pass down the ovaries to the oviduct and receptaculum seminis they become elongated and a single one completely fills the transverse diameter of the duct, and its length is twice the diameter. The mature egg is 80 /.i to 100 j.i long. The ends are rounded, and the form is somewhat curved so that it resembles a bean. The protoplasm in the young and transparent egg gradually becomes granular, and in the mature eggs the coarse granules and fat globules are so numerous that the egg becomes opaque aud the nucleus is seen with difficulty. After the eggs are fertilized in the receptaculum seminis they move on toward the uterus aud vulva, but it is very rare that they escape from the body of the cyst, since that is usually inclosed on all sides by the surrounding tissue. The uterus becomes ruptured and the eggs become free within the body cavity. Segmentation of the egg fre- quently begins before the egg escapes from the uterus. The eggs may be found in several stages of segmentation while still within the uterus. The first division is usually such that the egg is divided into two unequal cells, though the difference is not very great. From this time the segmentation is somewhat unequal. These two primary unequal DISEASES OF COTTON. 315 cells represent the mother cells of two different groups of cells. The larger cell divides more rapidly and forms small cells which completely surround the group of larger cells that result from the slower segmen- tation of the smaller primary cell. The growing over of these smaller cells proceeds first down the convex side of the egg. This ectoderm layer of cells folds over the opposite end of the embryo, the endoderm cells. Thus the prostom is upon the ventral side of the egg — that is, the concave side — because the ectoderm cells on the concave side of the egg have grown but little. If at this stage we turn the egg so that we are looking directly at the concave side, the ectoderm cells will be in a boat-shaped mass, and within this are the endoderm cells. The pros- torn now begins to close by the growth and increase of the cells at the margin. This closure takes place more rapidly at the caudal end and advances toward the cephalic end, so that there is finally only a small opening through the ectoderm near the cephalic end on the ventral side. This at last becomes completely closed. Invagination now takes place at the cephalic end, and the mouth and oesophagus are developed. Some of these phases can be seen while the embryo is still of the same length as the egg, but it soou begins to elongate and becomes more slender, and at the same time coils around within the egg. Eventually it is coiled three or four times. When it reaches this stage it remains a day or two within the eggshell, when by writhing and twisting the egg membrane is ruptured and the embryo escapes. In doing so it molts for the first time. Larval stage. — The larva is 300 /* to 400 jj. long; it tapers gently to the cephalic end and gradually into the pointed caudal end. In this form it resembles what are called "vinegar eels." The larva? at the time of hatching are inclosed usually by the surrounding tissue in which the female cyst was inclosed. They make their escape either by the decay and disintegration of the tissues or by battering a place of escape with the use of their spear. With the spear they also now gain an entrance into fresh and uninjured portions of the roots. The plant not being able to expel the invader bends its energies to the repair of the injury, and the result is the formation of a gall, by the increase of the cells at the injured part, all of which lessens the energies of the plant normally directed to the production of leaf and fruit. The larva? wander through the tissue for a time, when they molt again and pass into the cystic state. Cystic state. — The larva? locate at various depths in the tissues, and the body now begins to enlarge or "swell "up, except at the ends. Almost before any increase in the size of the body takes place it becomes rigid so far as any voluntary motion of any considerable extent is concerned. It may be twisted or turned in very curious forms when this rigidity comes upon it. The enlargement begins close behind the muscular bulb of the oesophagus, and for a little time this part of the body is a little larger than the caudal part. Very soon the enlargement 316 THE COTTON PLANT. takes place all along the body as far as the anus at the hyaline por- tion, some little distance from the pointed end of the tail. The cyst is at tirst irregularly spindle-shaped, then clavate with a sharply pointed process, the tail, at the larger end. Up to this time it is very difficult to distinguish the sexes, but from this point they strongly diverge. The female cyst continues to enlarge until it readies the vesicular form which it possesses at maturity, while the male undergoes a second transformation and returns to the thread-like or anguillula form. Transformation of the male. — The body of the male at this time is of the same size as that of the cyst, very stout in proportion to its length. The first sign of the transformation is the slipping of the head from this end of the cyst. At the same time the thick body begins to elongate, narrow, and double up within the cyst wall. This is the third molt, and while it is undergoing this change it molts again, making four in all. The male continues to elongate until it is coiled three, four, or more times within the cyst, according to the length of the latter, which retains perfectly its former shape. During this trans- formation the sexual organs of the male have matured. It now breaks through the wall of the cyst and travels through the tissues until it reaches a female, when fertilization takes place, after which the male soon dies. A life cycle of the worm is completed in about one month. Therefore in favorable seasons there would be seven or eight successive genera- tions in a single year. When we consider the number of eggs each female is capable of producing (from 100 to 200), it will be seen that the worms must multiply with startling rapidity. The periods of trans- formation of different individuals do not altogether coincide, so that at almost any season we may find worms in all stages of development. The injuries produced by the presence of the worm cause distortions of the tissue elements, and in many cases so devitalize the tissues that putrefactive organisms set to work and produce extensive diseased areas. The large amount of nutriment taken also by the roots in the development of the galls lessens the product of the plant. The great- est injury to cotton, however, seems to appear when the disease is accompanied with the " frenching" organism (see p. 290). Great care should be used in transferring rooted plants which are liable to be affected by the nematode from one place to another. Occasionally some males were found which showed but a single testis. Since Heterodera schaehtii possesses but a single testis, it might be well to inquire whether that species were also present and whether they are associated in the same roots in some cases or whether there is a varia- tion in U. radicicola in the possession of paired and single testes. This, however, must be left to future investigators. THE INSECTS WHICH AFFECT THE COTTON PLANT IN THE UNITED STATES. By L. O. Howard, Ph. D. Entomologist, United States Department of Agriculture. * The cultivation of cotton in the United States has gone through a curious change with regard to the depredations of insects. Kot very long ago, according to statistics published by this Department, the average annual loss to the cotton growers from the work of a single species of insect amounted to 815,000,000, while in such seasons as that of 186S and that of 1S73 the loss far exceeded this amount. This insect was the so-called cotton caterpillar, or cotton army worm, or chenille, or cotton-leaf worm (the larva of Aletia argillacea Hiibn.). Down to the year 1881 the damage done by this insect so far exceeded that inflicted by any other species that other forms had received but little consideration. With the rapid extension of cotton culture in Texas, however, the so-called bollworm (the larva of Heliothis armujer Hiibn.) became a very prominent factor in the cultivation of the crop. This insect had always been known as an enemy of cotton in the other States, but the damage which it accomplished was greatly inferior to that of the cot- ton-leaf worm. Texas, however, speedily became noted for bollworm damage, and as early as 1879 this insect was recognized as the chief depredator on cotton. The subject of the cotton worm attracted very considerable atten- tion at au early date. Southern newspapers and magazines contained many articles, of which those by Dr. D. B. Gorhain, Mr. Thomas Affleck, Dr. D. L. Phares, and Mr. William Jones are the most important. Mr. Townend Glover, the first entomologist to this Department, inves- tigated the subject of the cotton worm in the fifties, and, in fact, the literature of the subject increased very rapidly after the destructive year 1817. After the close of the civil war a series of cotton- worm years fol- lowed, unfortunately just at the period when the loss of a part of the crop meant more to the people of the South than had been the case at any previous time. This attracted investigation to the subject, and it was not long before the use of paris green upon the crop was suggested, probably in the first place by Prof. C. V. Eiley, at that time entomologist to the State of Missouri. It was later advocated by Mr. Glover, representing this Department, and circulars were sent out from 317 318 THE COTTON PLANT. the Department in the year 1873, urging the use of this arsenical poison upon the crop. After the conclusion of the work of the United States Entomological Commission on the Migratory Locust, or "Hateful Grasshopper" of the West, the most prominent and destructive insect in tbe country de- manding investigation seemed to be the cotton worm. Under appro- priations from Congress the investigation of this insect was carried out by the United States Entomological Commission and the Division of Entomology of this Department, working independently.1 At just about the time when the commission finished its labors upon this insect the insect itself ceased to be a serious enemy to cotton. Since that season the damage which it has done has been compara- tively small, and at the present time the cotton worm is not feared by planters. That this state of affairs is due in part to the work of the official investigations can not be doubted, but to claim that it is due entirely to this work would not be justified. The widely disseminated knowledge of the life history of the insect and the consequent general adoption of the plan of early poisoning (i. e., before the destructive third generation of the caterpillars has made its appearance) in the more southern portions of the cotton belt have had much to do with the recent comparative immunity; but aside from this, the cotton worm is undoubtedly much less abundant and destructive than it was fifteen years ago, and as a consequence it no longer holds the first position among the insect enemies of the cotton crop. With the bollworm, however, we find a different state of affairs. This insect, as will be shown in a later portion of this article, does not depend upon cotton alone for its existence. It feeds upon many differ- ent plants. The cultural methods in vogue all through the South are peculiarly favorable to its development, and it is, moreover, a hardy insect, with comparatively few natural enemies. Further, it feeds for the greater part of its existence protected by some portion of the plant which it infests, and in a solitary way, so that it is not subject to the contagious diseases which are so fatal to many injurious insects of gregarious habit. This latter cause also operates against successful treatment, and the result is that bollworm damage has not diminished, but in fact has increased in many portions of our cotton-gro wing- territory. ■The results of these investigations are contained in three volumes: The first is a 144-page pamphlet published as Bui. No. 3 of the Entomological Commission, Janu- ary 28, 1880; the second a 500-page volume issued by the Department of Agriculture and entitled "Report upon Cotton Insects," etc., by J. Henry Conistock, submitted November, 1879; the third, the "Fourth Report of the United States Entomological Commission," by C. V. Riley, published in 1885, a bulky volume of about 600 pages, illustrated by 64 plates and numerous text figures. Although its last report was not published until 1885, the work of the commission was practically completed with the year 1881. Copies of this report may still be obtained through Members of Congress. INSECTS WHICH AFFECT THE COTTON PLANT. 319 As shown above, the bollworm was incidentally studied during the Government investigation of the cotton worm. Its life history was completely made out, and fairly satisfactory remedies were discovered and displayed in Comstock's report on cotton insects and in the Fourth Report of the Entomological Commission. JSTo general adoption of these remedies resulted, however, and as a general thing planters either undertook no remedial work whatsoever, or relied upon such exploded remedies as the use of trap lanterns and poisoned bait. As a result of the continued damage done by this species, Congress provided, in the appropriation bill for the fiscal year 1890-91, for a supplementary investigation by the Division of Entomology of this Department. At that time considerable attention was being given to the subject of con- tagious diseases of insects, and the Entomologist was instructed to conduct a thorough series of experiments in the direction of the prac- tical use of some such disease as applied to this insect. The work of a season in the field by an assistant in the division, F. W. Mally, who was helped by Dr. A. R. Booth, of Vicksburg, and by Mr. Nathan Banks, of the office force, resulted in the publication of Bulletin 24 of the division, in which it is shown that the best dependence of the cot- ton grower is upon the remedies recommended in the Fourth Report of the Entomological Commission, and that nothing practical is to be hoped for in the way of the encouragement of any disease. Congress renewed the appropriation the following year, and Mr. Mally continued his work. The results are summarized in Bulletin 29 of the division, which is entitled, " Report on the Bollworm of Cotton," and which was published iu the spring of 1893. It is a pamphlet of 73 pages, and con- tains a number of new but comparatively unimportant facts relative to the economy of the insect and the details of the experimental work. No new remedies of value were brought out. Aside from the cotton worm and the bollworm, the cotton plant can not be said to suffer seriously from the attacks of insects. Cutworms sometimes damage the young plants in the beginning of the season; plant lice occasionally cause the withering of the terminal leaves (also usually early in the season); there are several bugs which sting the young bolls, although never to any serious extent; grasshoppers some- times "rag" the leaves in Texas, and there are several leaf feeding cat- erpillars which appear later in the season, and in reality do little but remove the superabundant foliage and expose the bolls to the sun, causing earlier ripening, and consequently a beneficial rather than an injurious effect. We occasionally learn of a case of local and tempo- rary damage by one or another of several species of insects, such as the garden webworm, which injured young cotton growing in proxim- ity to garden crops in Texas, Arkansas, and Indian Territory a few years ago ; but these cases are rare, and do not deserve extended con- sideration. A serious exception to this general statement may in the future be 320 THE COTTON PLANT. found in Anthonomus grandis, a Mexican weevil which damages cotton bolls. This insect, down to the close of the season of 1894, was known to us only through a few specimens collected upon cotton bolls in Mex- ico some ten years since by Dr. Edward Palmer. During 1894, how- ever, we learned that the species had made its appearance in the State of Texas. It works in a peculiarly injurious manner, utterly destroying many bolls. The life history of the species was carefully investigated during 1S95, and the writer has published two circulars of information, which have been widely distributed among cotton planters.1 After tins brief summary it will be evident that the subject of insects injurious to cotton in the United States will be most conveniently han- dled under four main headings — (1) the cotton worm, (2) the bollworm, (3) the Mexican cotton-boll weevil, and (4) other cotton insects. THE COTTON WORM, OR COTTON CATERPILLAR. (Aletia argillacea Hiibn.) GENERAL APPEARANCE, HABITS, AND LIFE HISTORY. This insect is perfectly familiar to all cotton growers. The slender, bluish-green caterpillar with small black spots, and often with black stripes down its back, which loops when it walks and feeds voraciously on both upper and under surfaces of the cotton leaf, is to be found in cotton fields in the Gulf States all through the summer. It is gen- erally not noticed in the early part of the season on account of its insig. nificant numbers. Later, through the ragging of the leaves, it becomes noticeable, and in seasons of abundance the plant is entirely defoli- ated. Farther north the insect makes its appearance at a later date in the season, and there the caterpillars are not the offspring of hibernat- ing moths, but ot the moths of the first or second generation, which have developed in more southern cotton fields and have flown north with the prevailing southern winds. Late in the season moths of the fourth or fifth generation fly far to the north, frequently making their appearance in numbers about electric lights in Canada There is no absolute evidence of any other food plant than cotton, although many entomologists have surmised that the species has a northern food plant. The specimens seen in Canada have, however, in all probability flown north from cotton fields in the Carolinas, and perhaps even farther south. The egg. — The egg is bluish green in color and of a different shade from that of the leaf, so that it can be rather readily distinguished. It is flattened-convex in shape, with many parallel longitudinal ridges con- verging at the center above. It is found usually on the underside of the leaves and as a general thing toward the top of the plant. In the neighborhood of 500 eggs are laid by each female, sometimes several upon one leaf, but never in clusters. The eggs are laid at night, since the moth is a night flyer. The duration of the egg state A'aries some- what, according to the season. In midsummer the larva hatches in 1 U. S. Dept. Agr., Div. of Ent. Circ. 6 and 14, n. ser. INSECTS WHICH AFFECT THE COTTON PLANT. 321 Fig. 9.-Eg° a, top view of cotton worm moth: 6, side view — greatly en- larged (from Fourth Report U. S. Entomological Commission). Before reaching- full growth from three to four days after the egg is laid, but iu spring and autumn this period is very considerably lengthened. The larva. — After hatching from the egg, the young larva feeds at first upon the underside of the leaf, devouring simply the lower paren- chyma and not piercing through to the upper side until after the first molt. At first the larva is pale yellow in color, soon becoming greenish. The dark spots be- come more or less conspicuous after the first molt, and the characteristic markings, as shown in the figure, make their first appearance. After the second molt these markings become more conspicuous, and the insect takes on a distinctly greenish color, the black along the back varying among different individuals iu its intensity the caterpillar sheds its skin five times, and the duration of the cater- pillar stage is from one to three weeks. Early in the season the green color appears to predominate, while toward the fall the blackish cater- pillars are more abundant, although at any time during the season green and dark worms are seen together. Although the normal food of the caterpillar is the leaves, it will frequently gnaw the tender twigs and will even damage the bolls by eating into them in spots. This, however, generally occurs only when the worms are present in exceptional numbers and the supply of leaves becomes exhausted. We have re- ferred to the fact that the caterpillar is a looper, i. e., that it walks by bringing its hind prop legs up to the true legs, causing its back to arch up in a loop. Like the true loopers, or measuring worms, it has the habit of jerking itself some little distance when disturbed, and when it falls it usually sup- ports itself by a silken thread. It is some- thing of a cannibal, and when other food fails, or even rarely when leaves are abun- dant, it will feed upon smaller and feebler individuals of its own kind. In spite of its comparatively small size and slender form, this larva is, in fact, very voracious, and when occurring in num- bers the ruin which it accomplishes is complete. The chrysalis or pupa. — The caterpillar, having become full grown, never enters the ground to transform, although many planters have believed that this is the. manner in which the insect passes the winter. 1993— No. 33 21 b Fig. 10.— Cotton caterpillar: a, from side, 6, from above — twice natural size (from Fourth Report TJ. S. Entomological Commission). 322 THE COTTON PLANT. Ci i b Fig. 11. — Cotton worm moth: a, with wings expanded in flight ; l>, wings closed, at rest — natural size (after Riley). It spins a light silken web, forming an imperfect cocoon, usually within a folded leaf. It is frequently seen hanging quite naked upon the plant, but in such cases the leaf in which it was originally spun has been eaten away by other caterpillars. Its color is at first green, but in the course of an hour or so it changes to brown. The insect remains in this condition for a period varying from one week to thirty days. The adult insect.— The perfect insect or imago of the cotton caterpillar is a rather small moth of an olive-gray color, sometimes with a some- what purplish luster. Its wings expand from ]i to \l inches. The markings of the wings are indicated in the figure. The moth is a night flyer and hides during the day, starting up and flying with a swift, somewhat dart- ing motion when disturbed. After sunset it takes wing and flies about, laying its eggs or searching for food. It feeds, in fact, rather extensively, frequenting neighboring flowering plants and also the nectar glands of the leaves of cotton. Fruit, as it ripens, also attracts these moths, and is frequently seriously injured by them. The tongue or proboscis of the moth is curiously modified and fitted for piercing the skin and tissues of ripe fruit, as is shown in the figure. It is said that they are able to puncture hard green pears, the effect of the puncture being a discoloration of the skin for some distance around. The female begins to lay her eggs in from two to four days after lea vingthechrys- alis, and each individual lays from 300 to GOO eggs. With five consecutive and rapidly developed genera- tions the occasionally extra- ordinary numbers of the late broods are not to be wondered at. Number of broods or generations. — The observations of Mr. Schwarz in South Texas in 1879 show that at least seven, and probably even more, generations are produced there. Fully as many probably develop in Florida. The general belief in the South up to the time of the beginning of the cotton-worm investigation was that there were three generations only, since three "crops" of worms only were customarily observed. The early generations, however, were overlooked on account of their small numbers, and, in fact, in the northern portions of the cotton belt, the general idea was correct enough, since northward- flying moths in general do not oviposit in fields in this region until FlG. 12. — Prohoscis of cotton worm moth— enlarged; tip at right still more enlarged (from Comstock's report on cot- ton insects). INSECTS WHICH AFFECT THE COTTON PLANT. 323 comparatively late in the season. The moths hibernate, only in the extreme southern portions of the cotton belt, as will be shown in the next section, and begin to lay their eggs as early as March, or perhaps even earlier, in South Texas and Florida. Two generations are rapidly developed, and then, in these localities, a confusion of generations com- mences on account of the retardation of development in certain indi- viduals and acceleration in certain others. Moths from the end of March on are constantly Hying out from these points and, carried by the prevailing southerly winds, settle in more northern fields and stock a certain number of plants with eggs. Moths' developing from cater- pillars hatching from these eggs in turn stock the fields in which they have developed with a greater number of eggs, and a certain proportion of them fly farther north. In this way there is a progressive develop- ment all through the cotton belt and a somewhat varying number of generations in different localities. Under certain conditions, however, such as the early development of a very large brood in the far South, so many moths may be developed that there is a nearly simultaneous stocking of a very extensive region. The importance of ascer- taining the early presence of the worms, although in small numbers, from a remedial point of view is very great, and since it was conclusively shown that worms may be found in the fields in the Gulf States long before the so-called " first crop," planters have looked for them more carefully, and doubtless in many cases pos- sibly severe injury has been prevented by the poisoning of early worms. The moths of the last generation in seasons of cotton-worm abun- dance frerruently make their appearance in numbers far North. The moth is a very strong flyer, and, aided by the wind, has been known to occur abundantly in Canada, and has been observed in numbers far out at sea. During September it has been known to do very consider- able injury to peaches in Kansas and to ruin acres of cantaloupes as far north as Racine, Wis. Method of passing the winter. — The greatest difficulty was found in settling the question as to the manner in which this insect passes the winter, but it has finally been established that over the more northern portion of the cotton belt the species dies out every year, while in the more southern portions the moth hibernates and remains torpid in sheltered situations. There must also have been occasionally an ingress of moths from outside of the United States, say from the West Indies Fig. 13.— Cotton worm egg parasite (Triehogramma pre- tiosa): a, adult female, greatly enlarged; b, ovipositor; c. female antenna ; d, male antenna, still more enlarged (from Fourth Report U. S. Entomological Commission). 324 THE COTTON PLANT. Fig. 14. — Ohalcis flavipes, an im- portant parasite of the cotton caterpillar (froni Fourth Report TJ. S. Entomological Commis- sion) . or from Mexico or Central America. It was undoubtedly in this way that the species was first introduced into the United States, and such immigrations were probably of frequent occurrence down to compara- tively recent years. Professor Riley, writing in 1882, concluded that there is nothing more fully established than that the moth hibernates principally under the shelter of rank wire grass in the more heavily timbered portions of the South, and that these moths begin laying on the rattoon cotton when it is only an inch or so high. Only the excep- \ \ tional few survive, and this survival seems to be more common in the western part of the cotton belt than in the Atlantic States. PARASITES AND NATURAL ENEMIES. In the report by Professor Comstock, pub- lished in 1880, and in the Fourth Report of the United States Entomological Commission much space is devoted to the subject of the natural enemies and parasites of the cotton worm. They are very numerous, and without their aid the worms must have done infinitely more damage than they have accomplished; but, practically speaking, we need devote no space to their detailed consideration, as they can not be practically handled. Their increase can not be encouraged beyond the enforcement of general laws against the killing of insectivorous birds. A few of the most important of the predaceous and parasitic insects are shown in the accompanying figures. The little egg parasite, Trichogramma pretiosa (fig. 13), is one of the most important. Mr. Hubbard has recorded the fact that in Florida this one parasite almost entirely annihilated the fifth brood. At the beginning of the fourth brood about half of the eggs were destroyed by this insect. Of the eggs laid by the fourth-brood-moths, from 75 per cent to 90 per cent were parasit- ized, while of the eggs of the fifth brood the proportion destroyed by the parasite exceeded DO per cent, and out of the sixth brood care- ful estimates show that but 3 or 4 eggs out of 100 escaped. The external parasite of the caterpillar Euplectrus comstocJcii (fig. 15), is also another abundant parasite, while the other insects figured take almost as important parts in limiting the increase of the worm. As far back as 1847 Dr. D. B. Gorham found that nearly all of the chrysalids of the last brood of worms were destroyed by Pimpla conquisitor (fig. 16). From this fact he argued that the fields must be restocked by moths flying up from the South, and perhaps from the West Indies. It is a very curious fact that some twenty-five Fig. 15.— Skin of cotton caterpillar attached to the underside of cot- ton leaf hy silk spun about the pupae of Euplectrus comstockii — natural size (from Fourth Report U. S. Entomological Commission). INSECTS WHICH AFFECT THE COTTON PLANT. 32i years later Mr. A. E. Grote, studying the cotton worm in Georgia, was unable to find any parasites whatever, and from this fact argued that the insect was not a normal member of the Georgia fauna, but flew in every year, probably from the West Indies. REMEDIES. In the Fourth Report of the United States Entomological Commis- sion nearly 200 pages were given to the consideration of remedies and preventive measures. All the false ideas which had gained currency among planters were explained away, an extensive con- sideration of remedies against the insect in all stages was given, and the subject of machinery for the distribution of wet and dry poi- sons was most elaborately treated. The chapters on remedies in this report have resulted in great benefit to the agricultural community as a whole. The system of eddy-chamber or cyclone nozzles was here first treated, and modi- fications of these nozzles are now in active use in all parts of the world for the application of insecticides and fungicides to very many crops. Sev- eral elaborate y riding on horseliack through the cotton rows, dusting two rows at a time. A little practice will enable one to do this work very evenly, and care must be taken not to allow the sacks to touch the leaves when wet or the poison will not pass through. Wheu the sacks are freshly filled a very slight jarring will shake out a sufficient amount of the poison, but when nearly empty the pole should lie frequently and sharply struck with a short stick, or spaces in the rows will be missed. 1 Three machines were patented in 1890 — the Roach cotton-worm destroyer, pat- ented by the James P. Eoach Manufacturing Company, Vicksburg, Miss. ; the Rogers dry-poison distributor, patented by the Rogers Company, of Lagrange, Tex., and the Brown machine, manufactured and sold by L. M. Rumsey & Co., at St. Louis, Mo. All of these cost $50 or more each, and while all were probably capable of doing the work claimed for them the sales were small. One company has gone out of business, and the machines of the others have not been extensively used. Should a season of great caterpillar abundance come, either of these machines may be used to good advantage, or the smaller spray distributor described in the Fourth Report of the Entomological Commission, or a number of the spraying machines put upon the mar- ket of late years in the North may be adapted to field use in the South, 2 Mississippi Sta. Bui. 2. 328 THE COTTON PLANT. When used iu this way we have found it the best plan to use the poison without any admixture of flour, and if flour is to be added lighter cloth should be used in making the sacks. With a pole and sacks as described, one man and mule can poison from 15 to 20 acres per day. THE COTTON BOLLWORM. (Heliothis armiger Hiibn.) Unlike the cotton worm, this insect is by no means confined to Amer- ica, nor is it confined to cotton as a food plant. It is known in many other parts of the world, and it can not be surmised at the present time whether it has been carried from some one point or whether it is indig- enous over its extremely wide range. Its food plants vary in extraordi- nary degree. In this country it is one of the principal enemies of cotton, of corn, and of the tomato. The cotton bollworm, the corn earworm, and the tomato fruit worm are all the same species. In addition to these crops, it feeds upon peas and beans, tobacco, pumpkin, squash, okra, and a number of garden flowering plants, such as cultivated geranium, gladiolus, mignonette, as well as a number of wild plants. GENERAL AI*PEARANCE, HABITS, AND LIFE HISTORY. The egg. — The egg is a little larger than that of the cotton worm and more nearly round. It is nearly white in color and rather inclined to yellowish. Examined with a lens, its structure seems to be almost identical with that of the cotton worm. The eggs are laid upon all parts of the cotton plant, occurring most abundantly on the underside of the leaf. A few can be found upon the stalks, many upon the upper* surface of the leaves, some upon the involucre, and occasionally they are seen upon the stems of the boll or upon the leaf of the petiole. The eggs are laid just at twilight, and they hatch in from two days to a week. The larva. — When first hatched, the bollworm looks much like the cotton worm. It is rather darker in color, but also walks like a looper, or measuriug worm. It feeds at first near the eggshell, and then begins to wander away, crawling from one leaf to another, until a young bud or boll is found, into which it bores. Frequently several days pass in this search for a boll, and rarely the worm may reach full growth upon a diet of leaves. It is during this early, wandering, leaf-feeding exist- ence that the insect may be destroyed by arsenical poisons, as is true of the cotton worm. When the young worm enters the flower bud the invo- lucre flares open and the young bud or young boll finally drops. This "shedding" of cotton is, however, not caused by the bollworm alone. Other insects are concerned in the damage, and the flaring and dropping occasionally occurs when no insect injury can be found. A very con- siderable amount of damage may be done in this way, as a single young larva will travel from bud to bud, deserting each before it falls. The bud pierced just before opening is forced into premature bloom, but the worm usually feeds upon the stamens and pistil, rendering it DESCRIPTION OF PLATE IV. TRANSFORMATIONS OF COTTON BOLLWOKM. (Ueliothts armujer Hiibn.) Fig. 1. Egg ou underside of cotton leaf. Fig. 2. Larva one-third grown boring into square. Fig. 3. Entrance bole of young larva in square, with excremental pellets at edge of hole. Fig. 4. Nearly full-grown larva just issued from boll. Fig. 5. Full-grown larva ou leaf stem. Fig. 6. Pupa shown in center of underground earthen cell; cell shown in longi- tudinal section. Fig. 7. Adult moth, light variety. Fig. 8. Adult moth with dark forewings. Fig. 9. Adult moth iu resting position, wings slightly elevated, hind border of hind wings slightly showing. U. S. Dept of Agriculture, Expt. Station Bui. No. 33 Plate IV. Transformations of Cotton Bollworm. INSECTS WHICH AFFECT THE COTTON PLANT. 329 incapable of fructifying. As the bollworms grow they begin to vary greatly in general appearance. Full-grown worms may be found of almost every intermediate stage of color between light green and dark brown or rose. They may be unstriped and unspotted, or they may possess dark stripes or black spots. These color varieties are not caused by different food, since many variations occur in specimens feeding upon the same plant. Upon cotton the larger worms take the larger bolls, the young ones having confined themselves in the main to the flower buds and the newly formed bolls. They then practically progress downward, the young ones being found mainly upon the top crop, while the older ones bore into the older bolls of the middle crop, tbe bottom crop being seldom seriously damaged by this insect. Often a single worm will practically destroy several large bolls, and one instance is on record where 18 young bolls and many blooms and unopened flower buds have been destroyed by one not fully grown worm. The bollworm is not only a voracious plant feeder, but it is also a cannibal. Older worms feed upon younger ones, and it has often been known to eat the chrysalids of the cotton caterpillar. With au abun- dance of vegetable food at hand, the larger worms will seize upon their small brothers, biting through the skin and feeding upon the juices of the body. In ears of corn the remains of several young worms are often found, while the strong, large worm which has destroyed them is the only living occupant of the ear. The larva occupies from two weeks to a month in reaching full growth. The pupa, or chrysalis. — Unlike the cotton caterpillar, the bollworm enters the ground in order to transform. It forms an oval cell com- posed of particles of earth held together by a loose, gummy silk, or the pupa may be perfectly naked. It is of a light mahogany color, darker toward the head, and the duration of this state is from one to four weeks. The adult insect. — The adult insect of the bollworm is a moth about the size of the cotton-worm moth, but has a stouter body and is more extensively marked, as well as more variable in its markings. Its general color varies from a dull ochre-yellow to a dull olive-green. The fore wings have a rather dark band near the tip and the hind wings are also bordered with a darker band. The wing veins are lined with black and the fore wings have also several dark spots. There is great varia- tion in these markings, and they are intensified in some individuals and almost lacking in others. When the moth is at rest, the fore wings are slightly open, whereas in the cotton- worm moth they are closed in a roof-shaped manner. The moth flies normally about dusk, lays about 500 eggs, and is not a fruit feeder like the cotton-worm moth. During the day they hide in cowpeas and in clover, when these grow near the cot- ton field, and fly low with a quick darting motion when disturbed. About sunset they begin to feed upon the honey secreted by the cow- pea and blossoms of clover, as well as upon the nectar of the cotton 330 THE COTTON PLANT. plant and other honey-secreting plants. Mr. Mally speaks of seeing tbe moths eating at 3 o'clock in the afternoon, and Mr. Mullen states that he has noticed them feeding freely during all hours except the early morning hours, and during 1892 noted them particularly deposit- ing their eggs in broad daylight. Number of generations. — The average time occupied by the insect in its transformations from egg to the adult is about thirty -eight days. The number of annual generations is about five. In the cotton -growing States the worms of the first three generations feed usually in the corn fields. In fact, in choice of food plants, cotton seems to be secondary to corn. They feed upon corn by preference until this becomes too hard to be readily eaten. The worms of the first generation make their appearance the latter part of April or early in May, and feed almost exclusively upon the leaves and terminal buds of corn. The second generation, appearing iu early June, feed upon the tassels and forming ears of corn, while the third appears hi July and feeds upon the hard- ening corn. When the fourth generation appears, the corn has become too hard for appropriate food and the moths, therefore, fly to neighbor- ing cotton, which carries at that time plenty of tender young bolls. A few worms will have been found upon cotton before this time and will have fed upon the leaves and flower buds only in the absence of bolls. Others will have been found upon tomatoes, if these are grown upon the plantation, while still others have been feeding upon cowpeas. As a genera] thing bollworms are seen in force upon cotton about the first of August, and usually these individuals belong to the fourth genera- tion. The fifth generation makes its appearance about the middle of September, and about the middle of October, or even earlier, the cater- pillars enter the ground for transformation to pupa?. Hibernation. — The bulk of the bollworms hibernate in the pupa state underground. In a warm fall the moths have been known to issue during the month of November, and Mr. Mally has shown that fre- quently a few moths hibernate. These hibernating moths appear and begin laying eggs much earlier than the moths which issue from over- wintered pupse. This results in something of a confusion of generations the following season, and at Shreveport, La., Mr. Mally found a series of small broods along with the more or less regular large ones, a sixth generation of worms appearing a little later in .the fall and hibernating in the pupa state. In evidence of this fact he adduces the finding of young bollworms as late as November 20. Young and old worms may, in fact, be found simultaneously after the middle of May. Mr. Mally's observations, however, were founded upon two seasons' observations only, and this state of affairs may be exceptional, particularly as the winter of 1890-91, when he made his observations, was unusually mild in Louisiana and the spring earlier than usual. In Arkansas four and five generations are found in the northern and southern portions of the State, while in southern Texas six generations and a partial seventh seems to be the rule. The determination of the time of the appearance of INSECTS WHICH AFFECT THE COTTON PLANT. 331 the several generations of moths for each differing locality is of very considerable importance, and can only be made by local observers. It is of importance in arranging for the trap-crop method of protecting cotton, which will be discussed under the head of remedies. NATURAL ENEMIES. The bollworm has by no means as many natural enemies as the cot- ton caterpillar. The latter insect feeds exposed upon the leaves, and is, therefore, subject to the attacks of predaceons and parasitic insects as well as birds. The bollworm, however, as a general thing, feeding in the interior of the cotton boll, or ear of corn, or fruit of tomato, or pea or bean pod, is not readily found. In fact, although birds have been noticed to feed upon it, it was long considered to be absolutely free from true parasites. Riley, however, bred a Tachina fly from the larva, and Hubbard reared the little egg parasite Trichogramma preti- osa from the bollworm eggs in Florida. The more recent investigations of Mally have resulted in finding four additional parasites. One of these is an egg parasite of the genus Telenomus. Another is a species of Limneria, while the other two are the common Eupleetrus comstochii How. and Chalets ovata Say, which are such abundant parasites of the cotton worm. The hairy or downy woodpeckers are frequent visitors of cornfields and have been seen to extract the worms from infested ears. REMEDIES. Lights for trapping moths. — This is one of the remedies which have been most often advised, and has been very extensively used in parts of the South, particularly in Texas. Schwarz, in 1879, found that a small trap lantern was extensively used by planters in the vicinity of Ilearne, Tex., and was inclined to think that some good was accom- plished by its use. The same year a large tomato grower at Mandarin, Fla., built large fires in his fields, with tbe evident result of greatly les- sening the number of worms in his crop, which had the previous year been almost entirely destroyed by them. In view of these facts, Mally, during his two summers' investigations, made extensive experiments with trap lights for the moths. He has carefully tabulated all the insects which were captured in this way.1 A few bollworm moths were caught, but these apparently by accident, and a thoroughly unpreju- diced conclusion from his experiments must be that the use of lights for attracting and trapping bollworm moths is without beneficial result. The other insects caught by the light were found to be about evenly divided between those which are beneficial and those considered inju- rious; but most of the insects called injurious are of no especial economic importance in the cotton region and should be omitted from consideration in forming conclusions. The use of lights, from a cotton- growing standpoint, is really a disadvantage, and money expended in this practice is without doubt entirely lost. U. S. Dept. Agr., Div. of Ent.Bul. 29. 332 THE COTTON PLANT. Poisoned stveets. — Together with the use of lanterns for attracting the moths, poisoned sweets have been recommended for many years. The first experiment of this kind of which we are aware was made by B. A. Sorsby, and is recorded in the Annual Report of this Department for 1855. He used vinegar and molasses in plates set upon small stakes in the cotton field, and his statement was that he captured from 18 to 35 moths to each plate for several days. Mally also experimented in this direction and found that a modification of this remedy is more or less effective. He advises the planting of a few rows of cowpeas as a trap bordering the cotton fields. They should be planted so late as not to reach the height of blooming before the destructive August brood appears. A portion of the row should be sprayed over every night with a mixture of 4 ounces of beer to 2 ounces of potassium cyanid solution. The moths will be attracted by this mixture and will be destroyed by it. The mixture dries readily and hence it applied in the afternoon will not result in the destruction of any day-flying bene- ficial insects. Poisoning the icorms. — The careful study which has been made of tbe natural history of the bollworm, particularly that by Dr. William Tre- lease, in Alabama, in 1880, shows that where arsenical poisons are applied for the so-called third brood of the cotton worm, about August 1, many bollworms are destroyed. It is about this time that many young worms are hatching from the eggs and feeding for a longer or shorter space of time on the leaves before entering the bolls. It was, therefore, thought at the time when Comstock's report on cotton insects was written that the poisoning for the cotton worm, which was so strongly recommended and which was so necessary under the con- ditions governing at that period, would largely reduce bollworm injury. In fact, as we have shown in our opening paragraph, the bollworm itself at that time was by no means such a factor in cotton growing as it is at the present time. With the great reduction of damage done by the cotton worm and the great increase of that done by the bollworm, however, poisoning for the cotton worm has become comparatively rare, and on account not only of the greater abundance of bollworms, but also of the consequent greater confusion of generations of this insect, and the fact that not more than half at the outside could be destroyed by poisoning at any one given time, this method has largely lost its former value as a bollworm remedy. Still, in seasons of great boll- worm abundance, and where early trap-crop methods have been neg- lected, poisoning, as recommended in an earlier paragraph, under the head of "The cotton caterpillar," will result in saving at least a part of the crop. Trap crops. — In the intelligent handling of trap crops the cotton planter will find by far the most efficacious preventive of bollworm damage. This suggestion is an old one. It was proposed by Sorsby in 1855, by E. Sanderson in 1858, and by Peyton King in 1859. It was recommended by Comstock after careful preliminary observations by INSECTS WHICH AFFECT THE COTTON PLANT. 333 Trelease in 1879, and Kiley in 1885 gives it at least equal rank as a remedy with poisoning. The complete development of the trap-crop system, however, rests upon the studies and recommendations made by Mally;1 and S. B. Mullen,2 of Ilarrisville, Miss., has written in a most practical manner relative to corn. Mally's recommendations are, in brief, when planting cotton leave vacant strips of 5 rows for every 25 of cotton. In these 5 rows, at the earliest possible time, plant 1 row with an early maturing sweet corn. It should not be drilled in too thickly, as a minimum number of plants and ears is desired. During the silk- ing period frequent careful examinations must be made as to the num- ber of bollworm eggs. As soon as no more fresh white eggs are found each morning, the silk ends of the corn should be cut away and burned or fed to stock in order to destroy the young worms and the eggs. A few eggs may also be found upon the leaves of the plants, and since no more growth is to be made the plants should be cut and destroyed. Then 3 more of the rows should be planted to dent corn at such a time as to bring the silking period about the 1st of July or a little later. Upon these rows very large numbers of eggs will be laid, but they should be allowed to mature in order that the natural enemies which parasitize the eggs and prey upon the larva? may not be destroyed. The crowded condition of the worms in the ears developed in these 3 rows will induce cannibalism to such an extent that the number of worms reaching maturity will be reduced to the minimum, and these can well be allowed to escape if the natural enemies are saved thereby. To trap these escaping individuals, however, the fifth and last row of the vacant strips should be planted to sweet corn at a time which will allow it to reach full silk about August 1, since the majority of the moths begin issuing again about that time. This last row should be carefully watched, and the corn should be cut and destroyed as soon as it appears that no more eggs are being deposited. Mr. Mally found that the corn produced by the second planting is likely to be large enough in quantity to pay for expense of cultivation and the sacrifice made by cropping the 5 rows in corn instead of cotton. Moreover, he thinks that if the first two plantings are well managed a number of the earlier broods of the bollworm will be so reduced that the August brood will not be capable of inflicting great injury, and therefore in the less-infested regions the third planting may be dispensed with. He further found that it was not necessary to crop the entire plantation with this 5 to 25 rows of corn to cotton. If 5 acres be planted in this way for every 50 acres of cotton, or even 5 acres of trap alternate for 75 or 100 acres, the crop of the entire plantation may be protected. The accompanying diagram (page 340) of a field of 1,050 acres will illustrate the proposed system. ^or the full details of Mally's experiments, see U. S. Dept. Agr., Div. of Ent. Bui. 29. 2 Insect Life, Vol. V, pp. 240-243. 334 THE COTTON PLANT. -r 1 Acres Corn AND OTTON CORN AN0 OTTON * to o a- 0 UO CO UJ Z 0 t- cc 0 <£ 1- O h- _ 0 < •- 0 , , up ° 0 3 xn ■—. «" -» UJ *- uj Z K P cc 0 o t- C/3 CO 0 1— CO 3 0 or z CC 5 0 CC z cc S3 O or z cc a Acres Corn AND OTTON Acres CORN AND OTTON H LO <~> h- h lC C-P CO -, UJ ^ UJ Z uJ Z uJ Z 0= P S £ CC O cd O o \- O (- O h- t ° uO '-' < £ up ^ acres Corn AND OTTON ACRES corn and OTTON uo o u3 Z cc 0 in tj uJ Z CC O 0 t- O H- < H < H UJ ss UJ Z or 0 0= P 0 (- 0 1- "J 2 S £* 0 Acr Cor AND OTTI CC q-Ol- LO O uO O cO cO « -» CO uJ Z uJ Z uj Z uJ Z cc a cc 0 CC O cc 0 O (- O (- O 1— 0 h- < H < 1- <£ H < 1- 0 S UP <-> CO 0 O CO — CO S3 3 03 ^ CO 0 CO z 0 Acre Corn AND OTTOI or z Qr or z cc ACRI CORN AND OTTO cc z cc u-i 0 3 h- 2 z cc 0 O r- 3i uO <_> h uj Z cc 0