RICE IN THE : ^ VARIETIES AND PRODUCTION n^

- -- m

-

CO

¡

Agriculture Handbook No. 289

Agricultural Reseorch Service U.S. DEPARTMENT OF AGRICULTURE

«OWUV, ,ou,S.ANA 70527-M29 y PESTICIDES PRECAUTION

Pesticides used improperly can be injurious to man, animals, and plants. Follow the directions and heed all precautions on the labels. Store pesticides in original containers under lock and key—out of the reach of children and animals— and away from food and feed. Apply pesticides so that they do not endanger humans, livestock, crops, beneficial insects, fish, and wildlife. Do not apply pesticides when there is danger of drift, when honey bees or other pollinating msects are visiting plants, or in ways that may contaminate water or leave illegal residues. Avoid prolonged inhalation of pesticide sprays or dusts; wear protective clothing and equipment if specified on the container. If your hands become contaminated with a pesticide, do not eat or drink until you have washed. In case a pesticide is swallowed or gets in the eyes, follow the first aid treatment given on the label, and get prompt medical attention. If a pesticide is spilled on your skin or clothing, remove clothing immediately and wash skin thoroughly. Do not clean spray equipment or dump excess spray material near ponds, streams, or wells. Because it is difficult to remove aîMi^ces of herbicides from equipment, do not use the same equipment for insecti- cides or fungicides that you use for herbicides. ^ Dispose of empty pesticide containers promptly. Have them buried at a sanitary land-fill dump, or crush and bury them in a level, i^çlatèd place. , '.. . , NOTE: Some States have réárictions on thenisp of certain pesticides. Check your State and local regulations. Also, because registrations of pesticides are under constant review by the Federal Environ- mental Protection Agency, consult your county agricultural agent or State Extension specialist to be sure the intended use is still registered.

FOLLOW THC LABEL

U.S. DErAITMENT OF ACIICUITUKE IN THE UNITED STATES: VARIETIES AND PRODUCTION

USDA,NatípnalAgrieünwaIl*«r NALBWg 10301 BaWmore BWO 'B^ville.MD 20705-2351 i

Agriculture Handbook No. 289

Agricultural Research Service U.S. DEPARTMENT OF AGRICULTURE

Washington, D.C. Revised June 1973

Fcr sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $$2.10, domestic postpaid; $1.75 GPO Bookstore Stock Number 0100-02717 PREFACE

This bulletin is a revision of United States Department of Agriculture Handbook 289, published in 1966. The revision includes the latest data on production, yield, varieties, soils and fertilizer practices, culture, and control of w eeds, diseases, and insects that were available at the time of revision. The information used to update the publication was taken from latest Agriculture Statistics (1969) and from current publications on the various phases of rice varieties and culture. Trade names are used in this publication solely for the purpose of providing specific information. Mention of a trade name does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture or an endorsement by the Department over other products not mentioned. Italic numbers in parenthesis refer to Selected References at end of each section.

CONTRIBUTORS

0. ROY ADAIE, research agronomist, Plant Genetics and NELSON E. JODON, research agronomist, Southern , Germplasm Institute, Northeastern Region, Agricul- Agricultural Research Service, Crowley, La. 70526 tural Research Service, Beltsville, Md. 20705 T. H. JOHNSTON, research agronomist, Southern Region, J. G. ATKINS, plant pathologist. Southern Region, Agricul- Agricultural Research Service, Stuttgart, Ark. 72160 tural Research Service, Beaumont, Tex. 77706 D. S. MIKKELSEN, professor of agronomy. University of 0. N. BoLLiCH, research agronomist, Southern Region, , Davis, Calif. 95616 M D. MILLER, extension agronomist. Agricultural Exten- Agricultural Research Service, Beaumont, Tex. 77706 sion Service, University of California, Davis, Calif. DONALD H. BOWMAN, agronomist, Delta Branch, Missis- 95616 sippi Agricultural and Forestry Experiment Station, D. E. SEAMAN, specialist in agronomy. Department of Stoneville, Miss. 28776 Agronomy and Range Science, University of Cali- N. S. EvATT (deceased), associate professor, AdM fornia, Rice Experiment Station, Biggs, Calif. 95917 university. Agricultural Research and Extension Cen- R. J. SMITH, JR., research agronomist. Southern Region, ter, Beaumont, Tex. 77706 Agricultural Research Service, Stuttgart, Ark. 72160 JAMES R. GIFFORD, entomologist, Southern Region, Agri- B. D. WEBB, research chemist. Southern Region, Agricul- cultural Research Service, Baton Rouge, La. 70803 tural Research Service, Beaumont, Tex. 77706 CONTENTS

Page Rice breeding and testing, etc.—Continued Page Introduction 1 Production of seed rice .__ 65 History of rice in the United States 2 Origin of high-quality seed rice 65 Distribution of rice in the United States 2 Classes of seed in a certification pro- Acreage, yield, and production of rice in the gram 66 United States 2 Cleaning, grading, and processing seed Selected references 5 rice 69 Distribution and origin of species, botany, and Standards for seed certification 70 genetics Q Selected references 71 Distribution of the species of Oryza and origin Soils and fertilizers 76 of cultivated rice 6 Types of soils used for rice production 76 Description and development of the rice plant 77 and classification of cultivated varieties 8 77 Description of plant 8 Texas 77 Development of plant 9 California 78 Classification of varieties 11 Chemistry of flooded soils 78 Genetics 11 Fertilizers 81 Selected references 19 Southern rice area 81 Rice breeding and testing methods in the united California 83 States 22 Micro nutrient deficiencies 85 History and objectives 22 Selected references 86 Cultural methods and equipment for breeding Culture 88 rice in the United States 23 Rotation or cropping systems 88 Breeding methods 27 Arkansas 88 Introduction 27 Louisiana 90 Selection 28 92 Hybridization 29 92 Irradiation 33 Texas 92 Breeding for agronomic characters 33 California 94 Testing for milUng, cooking, and processing Land leveling and seedbed preparation 94 qualities 38 Land grading and leveling 94 Milling quality 40 Seedbed preparation 96 Cooking and processing qualities 41 Seedbed preparation as related to method Breeding for disease resistance 44 of seeding 98 Blast 46 Construction of levees 99 Brown leaf spot 46 Seed and seeding 101 Narrow brown leaf spot 47 Seed quality 101 Straighthead 47 Source of seed 102 White tip 48 Seed treatment 102 Hoja blanca 48 Time of seeding 103 Description of varieties 49 Rate of seeding 105 Short-grain varieties 49 Method of seeding 106 Medium-grain varieties 50 Transplanting rice 110 Long-grain varieties 53 Irrigating and draining 110 Other kinds of rice 55 Amount of water required 110 Performance of varieties 56 Source of water 111 Choosing the variety 58 Quality of water 112 Varietal response to seeding date 60 Water temperature and oxygen content 113 Results of tests with older varieties 60 Water control methods 114 Results of tests with newer varieties and Water management 116 selections 61 Draining for harvest 119 iii CONTENTS

Culture—Continued Page Rice diseases—Continued Page Harvesting, drying, and storing 120 Major diseases—Continued Harvesting 120 Brown-bordered leaf and sheath spot 143 Drying and storing 124 Seedling blight 144 Selected references 128 Stem rot 144 Weeds and their control 135 Straighthead 145 Losses due to weeds and cost of weed control. _ 135 White tip 145 Problem weeds 135 Kernel smut 146 Losses from competition of specific weeds 135 Minor diseases 147 Weed control practices 136 Narrow brown leaf spot 147 Preventive, mechanical, and cultural con- Hoja blanca 147 trol 136 Kernel spots 148 Chemical control methods 137 Leaf smut 148 Herbicide drift 139 Selected references 148 Selected references 140

Rice diseases 141 Insects and their control 151 Major diseases 141 Rice water weevil 151 Blast 141 Rice stink bug 151 Brown leaf spot 142 Other pests of rice 152 Root rot 143 Selected references 154 RICE IN THE UNITED STATES: VARIETIES AND PRODUCTION

INTRODUCTION

By C. ROY ADAIR

Rice, a leading cereal crop in many countries, ond to Brazil in the . Other is grown on all . It often is considered leading rice-producing countries, outside of to be a tropical crop, althought it is grown in both and adjacent islands, are United Arab Republic the temperate and the tropical zones in , (Egypt) and Malagasy Republic (Madagascar) Asia, , , and South Amer- in Africa, and Italy and Spain in . ica, and in the southern part of Europe. About 93 percent of the world rice crop was produced Rice yields vary widely among the rice-produc- in Asia during the 5-year period ending in 1969 ing countries (table 1). Yields generally are (table 1). Only slightly more than 1 percent was much higher in temperate than in tropical produced in the United States during this period. zones, not only because of differences in climate The United States is, however, the leading rice- but also because of differences in cultural prac- producing country in North America and is sec- tices and in varieties grown.

TABLE I.—Rice acreage, production, and yield per acre for each and selected countries, averages for 5-year period, 1965-69

Continent and country Acreage Production Yield

1,000 acres Million pounds Pounds per acre Asia _ 287, 533. 7 558, 191. 4 1, 941. 3 India gg^ 720. 8 122, 798. 1 1, 368. 7 Japan g^ Qgg 4 37, 192. 0 4, 609. 6 23 ^25 8 20, 452. 3 I, 523. 4 Brazil 11' Qg2.* 8 15, 136. 4 °1, 365. 8 Peru 177 Q 632. 7 3, 574. 6 Africa 8^ 74I 4 14, 119. 5 1, 615. 2 Malagasy Republic 1 947 Q 3, 244. 1 1, 666. 2 United Arab Republic (Egypt) l' 197, 2 4, 961. 3 4, 144. 1 North America __ __ 3' ^32 o 11, 571. 2 3, 185. 2 Mexico ' 427* 4 873.3 2, 092. 2 United States 2 042 0 9, 067. 5 4, 440. 5 Europe _'."".".".'_'_' ' 826.* 6 3, 462. 1 4, 188. 4 Italy 359 4 1, 509. 8 4, 200. 9 Spain 15Q g 834.4 5, 533. 2 Oceania 200 0 580. 5 5, 805. 0 73 2 489.6 6, 688. 5 World total 317 144.5 610, 525. 0 1, 925. 1

Source: U.S. Department of Agriculture, Agricultural Statistics: 1966, p. 22; 1968, p. 22- 1969 p 22- 1970 p 21-22- 1971, p. 21-22. FAO, UN Prod. Yearbook: vol. 23, 1969, pp. 75-80; vol. 24, 1970, pp. 73-78. ' ' ' ' 1 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

History of Rice in the United States because of its fine texture, or a subsoil through which loss by seepage is small (6). These cli- Eice has been grown in the United States since matic and soil conditions prevail in the areas the latter part of the I7th century (2), Trial where rice is gro\v^n in the United States. plantings of rice were made in Virginia as early Eainfall and humidity during the growing sea- as 1609 (4). Apparently other plantings were son are comparatively high in the Southern States, made in the colonies along the South Atlantic so less irrigation water is required there than in coast from that time on, and ricegrowing was California (1), Irrigation water is supplied from firmly established in South Carolina about 1690. streams, from reservoirs where water is im- Until about 1890, rice in the United States was pounded in winter and early spring, and from grown principally in the Southeastern States, al- wells. To produce optimum yields, proper cultural though some was grown along rivers in the South practices must be followed. These practices include Central States. Experimental plantings were preparing a suitable seed bed, maintaining a uni- made in the prairie section of southwest Louisiana form depth of irrigation water, providing sufficient from 1884 to 1886 (7), and rice culture became soil nutrients for optimum growth, and controlling established in that area about 1888. Production insects and diseases and weeds and grasses. Good- then increased rapidly in that part of Louisiana, quality seed of adapted varieties must be used to and in the adjacent part of Texas. Some rice was maintain quality and to produce high economical grown along rivers in Arkansas in early years, yields. For safe storage, the rice must be harvested but it did not become an important cash crop in at the right stage and dried to the proper moisture the State until after 1904 (i^), when ricegrowing level. was started in Grand Prairie. Experimental plantings were made near Butte Creek in the Sac- Acreage, Yield, and Production of Rice in the ramento Valley in California in 1909 (3), and United States rice became established as a commercial crop in that area about 1912. Eice production has been In the 5-year period ending in 1969, Arkansas of considerable importance in the delta area of produced 24.67 percent of the total United States rice crop; Louisiana, 24.30 percent; California, Mississippi since about 1948 (2). 21.70 percent; Texas, 26.43 percent; Mississippi, 2.64 percent; and Missouri, 0.27 percent (8). The Distribution of Rice in the United States United States has been self-sufficient in rice pro- Although rice is probably the leading food duction since 1917, and now is (1972) the leading crop of the world, it is a major crop in the United rice-exporting country of the world. States in certain areas only. Eice production in Eice acreage in the United States (fig. 2) in- the United States is centered in the Southern creased from a 5-year annual moving average of States of Arkansas, Louisiana, Mississippi, and 301,000 acres in 1899 to 1,105,000 acres in 1922. It Texas and in California. It is the principal cash then declined until 1936. The 5-year annual mov- crop in many counties and parishes (fig. 1) (5). ing average acreage increased each year until 1955, Small amounts of rice also are grown in Missouri, when it was 2,106,000. The peak acreage of 2,550,- Oklahoma, South Carolina, and Tennessee. Some 000 was reached in 1954. The annual acreage then rice has been grown in each of the States in declined until 1957, when it was 1,340,000. Starting Southeastern United States. in 1958, acreage increased slightly each year until In the United States, satisfactory rice crops 1968, when it was 2,353,000. The acreage declined require (1) high temperature, especially rela- to 2,130,000 in 1969. tively high mean temperatures during the grow- Yield per acre (fig. 3) in the United States in- ing season; (2) a dependable supply of fresh creased from a 5-year annnual moving average of water for irrigation; (3) a terrain that is level 1,091 pounds per acre in 1899 to 4,381 pounds per enough to permit flood irrigation but that slopes acre in 1969. This gradual increase in yield per enough so that surface water can be readily acre has been brought about by improved cultural drained; and (4) soil that will hold water well practices, such as better rotations, weed control, RICE UN THE UNITED STATES

ARKANSAS ARKANSAS, CONT. CALIFORNIA LOUISIANA SECTION AND ACRES SOUTHWEST SECTION AND ACRES SECTION AND ACRES COUNTY Lafayette ^ COUNTY PARISH NORTHEAST Little River \ 2,060 SACRAMENTO VALLEY NORTH Clay 8,070 Miller Í Butte 58,206 Caldwell 333 Craighead 17,657 TOTAL ACRES Colusa 82,247 Catahoula 287 Crittenden 6,972 ARKANSAS 1970 .... 441,326 Glenn 44,857 East Carroll 2,480 Cross 35,541 Placer 4,084 Franklin 621 Faulkner 465 TEXAS Sacramento 9,383 Madison 1,513 Greene 5,380 SECTION AND ACRES Sutter 61,378 Morehouse 9,456 Independence 875 COUNTY Yolo 23,387 Ouchita 1,070 Jackson 20,718 EAST Yuba 15,390 Richland 846 Lawrence 8,390 Tensas 127 Mississippi 1,503 SAN JOAQUÍN VALLEY Chambers 43,676 West Carroll 985 Poinseff 39,021 Hardin 2,006 Fresno 14,911 RIVER Randolph 2,350 Jefferson 59,115 Kern 1,805 St. Francis 18,544 Liberty 37,023 Kings 290 Ascension 480 White 1,160 Newton 429 Modera 580 Pointe Coupée 595 Woodruff 20,705 Orange 2,777 Merced 6,152 TECHE San Joaquín 6,990 CENTRAL CENTRAL Stanislaus 1,961 Avoyelles 2,880 Arkansas 77,227 Brazoria 49,735 Tulare 660 Iberia ...» 6,300 Clark 563 Ft. Bend 23,085 Lafayette , 10,020 Hot Springs 478 TOTAL ACRES Galveston 8,822 Rapides 750 Jefferson 18,200 CALIFORNIA 1970 .... 332,281 Harris 28,144 St. Martin 4,100 Lee 8,412 Waller 14,352 MISSISSIPPI St. Mary 3,450 Lonoke 39,363 WEST Monroe 14,690 COUNTY ACRES SOUTHWEST Perry 1,010 Austin 3,223 Bolivar 21,902 Acadia 93,571 Phillips 5,126 Bowie 564 Coahoma 2,074 Allen 24,555 Prairie 40,580 Calhoun 6,882 De Soto 1,283 Beauregard 4,715 Pulaski 2,295 Colorado 36,304 Humphreys 2,131 Calcasieu 67,469 SOUTHEAST Jackson 29,625 Isaquena 108 Cameron 12,419 Lavaca 5,985 Leflore 3,740 Evangeline 45,397 Ashley 6,521 Matagorda 46,804 Quitman 930 Jeffers n Davis 98,018 Chicot 10,065 Victoria 4,570 Sharkey 1,063 St. Landry 17,290 Desha 14,106 Wharton 64,548 Sunflower 4,528 Vermillion 115,450 Drew 4,526 TOTAL ACRES Tallahatche 484 Lincoln 8,753 Täte 121 TOTAL ACRES TEXAS 1970 467,669 Tunica 3,454 LOUISIANA 1970 525,177 Washington 9,708

TOTAL ACRES MISSISSIPPI 1970 51,526

FIGURE 1.—Distribution of United States rice acreage in principal producing States. Source: (8). AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

2,500 Production of rice (100-pound bags) in the United States in 1718 was estimated to be over

2,250 — 79,000 bags of rough rice (5). Production in- creased to nearly 3.5 million bags in 1849. It

S> 2,000 — then declined to about 637,000 bags by 1871. By 1899, production had increased to 3.4 milHon bags. 1,750 The 5-year annual production increased to 19.4 million bags in 1922. Production declined until 1,500 — 1926 and then increased slightly. However, it remained within the range of 14 to 20 million 1,250 bags until 1936. The annual production fluctu- ated from year to year but with an upward trend 1,000 during the period from 1937 to 1954 when pro- duction reached a high of 64.2 million bags. 750 — Annual production declined until 1958 and then

500 — 4,500

250

1899 1909 1919 1929 1939 1949 1959 1969 YEAR

FIGURE 2.—Harvested acreage of rice in the United States 3,500 (5-year annual moving average), 1899-1969.

.5 3,000 -— and irrigation and fertilization practices; better machinery, which led to improved and more time- ly field operations; better methods of controlling diseases and insects ; and improved varieties. Av- 2,500 -— erage yields have fluctuated from year to year, with a long-term upward trend. However, there have been short periods when average yields de- ±! 2,000 — clined. The 5-year annual moving average yield per acre increased gradually from 1899 to 1940,

but six times during this period the 5-year aver- 1,500 age yield was lower than that of the previous year. Yields declined from 1941 to 1945. Labor and equipment were scarce during the war years, 1,000 — and the acreage was expanded to include some fields with poor soil. These factors may have accounted for the lower yields in this period. The

5-year average yield per acre increased each year 1899 1909 1919 1929 1939 1949 1959 1969 from 1946 through 1969. Higher rates of ferti- YEAR

lizer application and other improved cultural FIGURE 3.—Yield of rice per acre in the United States (5- practices increased yields during this period. year annual moving average), 1899-1^69. RICE IN THE UNITED STATEIS increased each year until 1968 when the annual Selecfed References production was 107.6 million bags. The production in 1969 was 91.9 million bags and the 5-year an- (1) ADAIR, C. R., and ENGLER, KYLE. "" nual moving average was 88.6 million bags (fig. 4). 1955. THE IRRIGATION AND CULTURE OF RICE. Ifl Water, U.S. Dept. Agr. Yearbook of Agr., pp. 389-894. (2) MILLER, M. D., and BEACHELL, H. M. 1962. RICE IMPROVEMENT AND CULTURE IN THE UNITED STATES. Adv. in Agrou. 14: 61-108. (3) CHAMBLISS, C. E. 1912. A PRELIMINARY REPORT ON RICE GROWING IN THE SACRAMENTO VALLEY. U.S. Dept. AgT., Bur. Plant Indus. Cir. 97,10 pp. (4) GRAY, L. C, and THOMPSON, E. K. 1941. HISTORY OF AGRICULTURE IN THE SOUTHERN U.S. TO 18 60. Carnegie Inst. Wash. Pub. 430. [Reprinted in 2 v.] Peter Smith, New York. (5) HOLMES, G.K. 1912. RICE CROP OF THE UNITED STATES, 1712^1911. U.S. Dept. Agr., Bur. Statis. Cir. 34. 11 pp. (6) JONES, J. W., DOCKINS, J. O. AVALKER, R. K., and DAVIS, W. C. 1952. RICE PRODUCTION IN THE SOUTHERN STATES. U.S. Dept. Agr. Farmers' Bui. 2043, 36 pp. (7) KNAPP, S.A. 1899. THE PRESENT STATUS OF RICE CULTURE IN THE UNITED STATES. U.S. Dept. Agri. Bot. Div. Bui. 22, 56 pp. (8) RICE MILLERS ASSOCIATION. 1970. RICE ACREAGE IN THE UNITED STATES 1970. Rice Jour. 73(8) : 10-11. (9) UNITED STATES DEPARTMENT OF AGRICULTURE. 1970. RICE SITUATION. U.S. Dept. Agri. Econ. 1899 1909 1919 1929 1939 1949 1959 1969 Res. Serv., RS 15, 33 pp. YEAR (10) ViNCENHELLER, W. G. FIGURE 4.—Production of rice in the United States (5- 1906. RICE GROWING IN ARKANSAS. Ark. Agr. year annual moving average), 1899^1969. Expt. Sta. Bui. 89, pp. 119-129. DISTRIBUTION AND ORIGIN OF SPECIES, BOTANY, AND GENETICS

By C. ROY ADAIR and NELSON E. JODON

Distribution of the Species of Oryza and Five species were recognized by Chatterjee {IS) Origin of Cultivated Rice but not by Tateoka {6J4) so they were not listed in table 2. These are 0, granúlala Nées & Am. Species of Oryza have been reported from all ex Hook, f.; 0, perennis Moench; 0. sativa L. continents except Europe and from many of the var. fatua Prain; 0, stapfii Eoschev.; and 0, larger islands. Eoschevicz {5S) reported a com- subulata Nees. Four species were recognized by prehensive study of the genus, and he concluded Tateoka {6^) but not by Chatterjee {IS) so they from his study that there were 20 species of were not listed in table 2. These are 0. angusti- Oryza, Chevalier {13) reported a similar study folia Hubbard; O, iarthii A. Cheval.; 0. longi- in which he recognized 22 species. Chatterjee glumis Jansen; and 0. rufipogen Griff. The {IS) later summarized the information on Oryza chromosome number of the species of Oryza as and listed 23 species. Tateoka {6J^) summarized reported by Kihara {35) and the distribution as the information on species of Oryza in 1963 and reported by Chatterjee {H) also are shown in recognized 22 valid species. The 18 species that table 2. were recognized by both Chaterjee {IS) and Eoschevicz {53) divided the species of Oryza Tateoka {6J^) are listed in table 2. into four sections on the basis of morphologic

TABLE 2.—Species, chromosome number, and distribution of Oryza

Chromo- Species some No. Distribution (2n)

1. 0. alia Swallen 24 and 48 South America and . 2. 0. australiensis Domin 24 Australia. 3. 0. brachyantha A. Cheval. & Roehr 24 West Tropical Africa and Central Africa. 4. 0. breviligulata A. Cheval. & Roehr 24 West Tropical Africa. 5. 0. coardata Roxh 48 India and Burma. 6. 0. eichingeri Peter 48 East Africa. 7. 0. glaberrima Steud 24 W^est Tropical Africa and Central America.^ S. 0. grandiglumis (Doell) Frodoehl 24 and 48 South America. 9. 0. latifolia Desv 48 Central America, South America, and . 10. 0.meyeriana (Zoll.cfc Mor.) Baill 24 Java, Borneo, Philippines, and Siam. 11.0. minuta Presl 48 Malay Peninsula, Phihppines, Sumatra, Java, and Borneo. 12.0. officinalis Wall, ex Watt 24 India and Burma. 13. 0. perrieri A. Camus Madagascar. 14. 0. punctata Kotschy ex Steud Northeast Tropical Africa. 15. 0. ridleyi Rook, f 48 Malay Peninsula, Siam, Borneo, and New Guinea. 16. 0. sativa L 24 and 48 India and Indo-China. 17. 0. schlechten Pilger New Guinea. 18. 0 tisseranti A. Cheval Central Africa.

1 P.I. 269630. Collected in El Salvador by H. M. Beachell and identified by Eugene Griffith, Plant Science (formerly, Crops) Research Division, Agricultural Research Service, Beltsville, Md., and J. R. Swallen, Smithsonian Institution Washington, D.C. 6 RICE IN THE UNITED STATElS characters and geographic distribution. Kihara that they occur natually in areas far removed from {S5) modified this grouping to include species not the center where rice cultivation originated. reported by Roschevicz. Many investigators have studied and discussed In 1963, a committee ^ reviewed the classifica- the origin of the genus Oryza and of cultivated tion and nomeclature of Oryza and recommended rice {13, U, 15, 35, J^S, J48, 50, 53, 55, 56, 67, 70), that standards be adopted and used uniformly. It is generally concluded that the original ances- This committee recognized 19 valid species of tral species may no longer exist and that present Oryza, Seventeen of these species are listed in varieties evolved through progressive stages from table 2. The species not recognized by this com- known wild species {56), mittee that is listed in taible 2 is Ö. grandiglumis. Nandi {44) and Sakai {54) proposed that a The other two species recognized by this com- species with a haploid number of five chromo- mittee are 0, angustifolia and 0. longiglwfnis. somes was the ancestor of Oryza, This proposal The committee also believed that certain aspects was rather widely accepted. Nandi (^^) observed of taxonomy in Oryza are uncertain. These are: that in the somatic complement having 24 chro- (1) The relation and nomenclature among the mosomes there were two members of eight types taxa commonly designated as Ö. sativa var. fatua and four members of two types. The maximum (or /. spontanea Roschev.) and 0, perennis association in second metaphase was two groups of (Asiatic, American, and African) subspecies and three and three groups of two. It was concluded varieties; (2) the relation of the form commonly from this observation that the haploid genome of designated 0. stapfti to 0. glaherrima and 0. hre- the present 0, sativa is composed of two original mliguLata; (3) the relation between 0. granulata five-paired species belonging to two different f. and (9. meyeriana; (4) the relation between 0. ancestral genomes in which two chromosomes were alta and 0, grandiglumis; and (5) the status of duplicated. the taxa previously designated 0, uianghensis Later Shastry, Rao, and Misra {60) identified Chev. and 0. malampuzhaensis Krish. and Chand. 12 pachytine bivalents in a strain of 0, sativa, The committee also believed hat the form com- based on their length and arm ratios. This finding monly designated 0, suhidata should be excluded seems to cast considerable doubt on the validity of from Oryza and should be recognized as Rhyn- the supposition "that 0, sativa is a secondarily choryza suhulata (Nees) Baill. balanced allo-tetraploid which originated through Chatterjee (i^) reviewed the literature on the hybridization between two different five-paired origin and distribution of wild and cultivated species in which two chromosomes were duplicated, rice. He concluded that the eastern part of India, probably due to meiotic irregularities in the hy- Indo-China, and part of China could be considered brid. This followed by a subsequent doubling of the area where cultivated rice {Oryza sativa) the chromosomes attained the secondary balance of originated. Chatterjee further concluded, as did n = 12, the present existing number of 0, sativa)'^ Roschevicz {53), that for the genus as a whole or {44)^ the section Sativa Roschev., the center of origin is Sampath and Rao {56) "inferred that Oryza Africa. He was of the opinion that 0, alta^ 0, perennis is the ancestral form of cultivated , australiensis^ 0. brachyantha^ 0. eichingeri^ 0. having given rise to O, sativa in Asia and 0, gla- herrima in Africa by human selection." These grandiglumis^ 0. latifolia^ 0, wdnuta^ 0. perrieri^ authors were of the opinion that 0, hreviligulata 0. schlechteri^ O. suhulata^ and 0. tisseranti had and 0, sativa var. fatua are of collateral descent little part in the ancestry of cultivated rice. This from O. perennis. The present types that are clas- view seems to be based either on the fact that these sified as 0, sativa var. fatua or spontanea show species do not cross with 0. sativa or on the fact morphologic differences that may be due to genetic transfer from cultivated rice. From this it is in- ^ This committee was appointed at the Rice Genetic ferred that these wild forms are derived from Symposium convened at the International Rice Research hybrids with cultivated rices and are not the pro- Institute, Los Banos, the Philippines. The members were genitors of cultivated rice. Instead it has been S. Sampath, India; T. Tateoka, Japan; and M. T. Hen- derson, United States. strongly suggested ( 55, 56, 70) that an interme- 8 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

diate type may be the immediate ancestor of O, There are many morphologic differences be- sativa. tween typical japónica, indica, and bulu varietés (table 3). Varieties that are known to be prog- Description and Development of the Rice enies of hybrids are intermediate in many Plant and Classification of Cultivated respects. Varieties Description of Plant Cultivated varieties of 0, sativa are divided into The principal parts of the rice plant are a groups on the basis of several characters. The fibrous root system, culms, leaves, and panicles. principal division is on the basis of sterility of the The fibrous root system extends outward and hybrid. Kato and others {SS^ SI^) observed that in downward from the base of the plant. Adventi- crosses of certain varieties from the temperate zone tious roots that arise from the lower nodes of the and certain varieties from tropical areas, the hy- culms are finely branched. The extent of the root brids had a high percentage of sterile florets. The system and the size of the roots vary with variety hybrids between varieties within these two groups and type of culture. were as fertile as self-pollinated parent varieties. The culm and leaves develop from the plumule. They proposed that the temperate zone varieties The culm consists of the nodes, which have solid be named japónica and the tropical zone varieties centers (septum), and the internodes, which are indica. Kato and others {Sl^) observed no differ- hollow. The culms of most varieties of Oryza ence between japónica and indica varieties in sativa are erect or ascending, although there are chromosome number and behavior or in pollen procumbent types. Varieties range in height formation. However, in the hybrid, pollen forma- from less than 15 to more than 96 inches (37 to tion was abnormal. Serological investigations 240 centimeters) although varieties grown in the showed differences between japónica and indica United States range from 36 to 54 inches. The varieties. culms vary in diameter at the base from 5 to 15 Later, Terao and Midusima {^66) noted that an- millimeters. The number of nodes in the culm other group of varieties, principally from tropical ranges from 13 to 16 {32)^ and the number is islands of Southeast Asia, were intermediate. This significantly correlated with length of growing last group is referred to as bulu. season. Usually, four internodes elongate; the It appeared from the earlier studies that the upper internode (peduncle) is usually the long- varieties could be divided into three distinct est, and it bears the panicle. groups based on sexual affinity. Mizushima {89)^ The leaves of the rice plant are flat and range however, determined that this was not true and from 7 to 20 millimeters wide. The coleoptile that sexual affinities among varieties from these usually is considered to be the first leaf. It has groups varied gradually from one extreme to the stomata but few chloroplasts. The mature foliage other. leaf consists of the sheath at the base, which sur- Oka (^5) compared 147 varieties from widely rounds the culm for some distance; the blade, different geographic areas on the basis of a number which is set at an angle with the sheath; the of morphologic and physiologic characters and by ligule; and the auricles. The junction of the observing sexual affinity in hybrids. From these sheath and blade often is called the collar or studies, he placed the 147 varieties tested into in- junctura. The swollen zone at the base of the dica (continental) and japónica (insular). He sub- sheath where it joins the culm is the pulvinus. divided the japónica group into "tropical-insular" The panicle of most varieties is fairly dense and "temperate-insular." and drooping. There is, however, much inter- In a review on reports of sterility in hybrids varietal variation, since the panicle ranges from between indica and japónica varieties, Shastry, oi)en to very compact and from erect to drooping. Kao, and Misra {60) explained sterility on the From one to threi^ or sometimes more branches basis of chromosone structural differences, lethal arise alternately or somewhat in whorls at each genes, inversions, paracentric inversions, genie node of the peduncle. The rachilla bears the mutations, and cryptic structural hybridity. spikelet, and each rachilla arises on the same side RICE IN THE UNITED STA'TElS

TABLE 3.—Characters of japónica, indica, and bulu rice varieties

Character Japónica Indica Bulu

Grain shape Short Long- Large. Second foliar leaf: Length do do Long. Angle Small Large Small. FoUage color Dark green Light green Light green. Culm: Stiffness- Stiff Not stiff Stiff. Erectness Upright Spreading Upright. Length Short Long Long. Flag leaf: Angle Medium Upright Medium. Shape Narrow and short Narrow and long Broad and long. Degree of emergence of upper Medium Well emerged Not emerged. node. Number of tillers Many Many Few. Pubescence Pubescent Variable Variable. Awns None None Many. Panicles : Number Many Many Few. Weight Heavy Light Heavy. Length Short Long Medium. Density Dense Medium Do. Branching of rachis Few do Many.

Source: Nagai UO).

of the rachis. Each rachilla usually has a single Development of plant spikelet, although some varieties may have two or Rice seed germinates rapidly when moisture, more spikelets on a rachilla. The spikelet is temperature, and oxygen are optimum. The em- laterally compressed, is one flowered, and articu- bryo, w^hich consists of the organs that develop to lates below the outer glume. produce the rice plant, lies in a slanting position The two outer glumes usually are short, that is, at the base of the grain on one side of the endo- less than one-third the length of the lemma, al- sperm. The endosperm contains the stored food though there are types that have glumes as long that nourishes the developing seedling until the as the lemma or longer. The lemma (inferior or roots have developed sufficiently to obtain nu- lower flowering glume) is rigid, keeled, and three trients from the soil. nerved, and has two additional thin, marginal At germination, the coleorhiza pushes through nerves that can be seen only in section. The the pericarp, leaving a cavity. The primary root apiculus or tip of the lemma is sometimes pro- soon elongates, fills the cavity, and then pushes longed to form the awn. The palea (superior or through the coleorhiza. The root extends upward upper flowering glume) is similar to the lemma, for a short distance and then turns down. About but narrower. It has two nerves near the margin this same time, the coleoptile emerges and rapidly and a thin midnerve that can be seen only in sec- elongates. The epicotyl below the coleoptile also tion. The flower is composed of a one-carpellate elongates. Elongation varies with the depth the pistil with a bifurcated, plumose slender style; seed is planted. It elongates enough to bring the and six stamens with yellowish, four-lobed, two- base of the coleoptile to the surface of the soil. celled anthers. The principal parts of the cary- When the coleoptile emerges from the soil, it opsis (brown rice) are the embrio and endo- splits on the side opposite the scutellum, and the sperm and the covering tissues. foliage leaves soon appear. Usually in about 2 10 AGRICULTURE PIANDBOOK NO. 2 8 9, U.S. DEPT. OF AGRICULTURE days, two adventitious roots start to develop in perceptible when the panicle primordium is 5.0 the meristematic zone on the side opposite the millimeters long. The panicle primordium then scutellum. A third adventitious root may appear elongates, and the spikelets are differentiated later at the coleoptile divergence opposite the two progressively from the tip down to the base of the older ones. Lateral roots soon develop from the panicle. Spikelets in the tip and middle of the primary adventitious roots {71), panicle have reached maximum length by the The main axis of the plant, often called the time the panicle has completed its elongation, primary tiller, is differentiated rapidly. The which is 15 to 20 days after formation of panicle meristematic region is at the base of the inter- primordium. The spikelets at the base of the nodes; this is typical of all grasses. With three panicle complete their elongation by the time the early (130-day) united States varieties, tillering panicle emerges from the sheath {4-9), started within 3 weeks after germination, and all Similar results were reported for a variety from tillers that produced panicles were formed within the Philippines {30), Anthesis starts the first a 3-week period (^). Japónica varieties started day of emergence of the panicle. It begins with to tiller about 14 days after transplanting and the flowers at the tip of the panicle and continues tillering was completed in 23 to 25 days {32), progressively at the tip of each branch of the Indica varieties started to tiller about 14 days panicle. The greatest number of flowers open after transplanting, and all tillers were initiated the second or third day after the panicle emerges. in 3 to 5 weeks from this time, depending on the Anthesis occurs over a period of 6 to 10 days, length of growing period of the variety H9), varying with weather and variety. It usually oc- Tillers develop from buds in the axil of the node curs from midmorning to shortly after noon. The and the arrangement is alternate {S2^ Jf9), Japón- time differs with variety, location, and weather ica varieties {32) develop secondary tillers from (^, 52), Pollen is shed just before or at the time the buds at nodes 3 or 4 to 10. Tillers from node tlie flower opens. 10 often do not produce panicles. Tertiary tillers The formation and development of the embryo may develop from buds at nodes of the secondary of a japónica variety {65) and of a variety from tillers. the Philippines {30) has been described. The Panicle formation starts when all nodes have process is similar in both varieties. Juliano and been formed. The length of the period from Aldama {30) reported that "development of the seeding or transplanting to the starting of panicle megasporange, megaspore mother cell, and em- formation varies with the length of growing bryo is normal and follows very closely those re- period of the variety. The period up to forma- ported by other investigators. The embryo sac tion of the panicle primordium constitutes the is of the normal octonucleate type." These inves- "vegetative" stage. It is this period that accounts tigators also observed that "development of the for the variation in length of growing period microsporange is normal and follows those re- among varieties. Formation of the panicle pri- ported in many angiosperms" ; that "division of mordium starts about 50 days after seeding for the microspore mother cell is successive"; and very early United States varieties and about 100 that "long before anthesis the young microspore days for later United States varieties. The period contains two coats, a thin peripheral cytoplasm, from formation of the panicle primordium to wherein a single nucleus is embedded, and a single emergence of the panicle from the sheath ranges germ pore." Double fertilization occurs about 12 from 24 to 31 days for indica varieties {Jf9) and hours after anthesis {65), In a japónica variety, from 28 to 36 days for japónica varieties {6), the caryopses reach maximum length in about 12 For indica varieties, the panicle primordium days after flow^ering, maximum width or breadth can be noted when it is 0.125 to 0.25 millimeters ( dorso ventral diameter) in about 22 days, maxi- long. By the time it is 0.5 millimeters long, mum thickness (lateral diameter) in about 28 "hemispherical excrescences are observed at the days, and maximum dry weight in about 35 days base" {49), The hemispherical excrescences rep- {37), Similar results were obtained for United resent the branches of the panicle. Eudiments of States varieties; air-dry kernel weight and per- the spikelet on either side of the basal axis are centage of germination were maximum about 35 RICE IN THE UNITED STATES 11 days after the first panicles emerged from the Much work has been done to devise objective meth- sheath {61). ods of classifying rice varieties. Previous work was summarized and a proposal for clasifying rice vari- Classification of varieties eties was formulated by a special committee in Cultivated varieties of rice belong to the genus India {21). The committee proposed that rice vari- Oryza L., tribe Oryzeae, and family Gramineae. eties be classified on the basis of qualitative and Most cultivated varieties are in the species Oryza quantitative characters. Nagai {Jfi) reviewed sys- sativa L., although varieties of the species 0, gla- tems used to classify rice varieties. These systems herrima are cultivated in Africa. 0, sativa is an were followed rather closely in the Food and Agri- annual; but when moisture and temperature are culture Organization, "World Catalogue of Ge- optimum and in the absence of disease, plants have netic Stocks—Rice," {1) and subsequent supple- survived and produced grain for 20 years or more. ments. The classification system used in the United All varieties of rice grown in the United States States includes the items in these systems with are in the species 0, sativa L. The commercial certain modifications and additions. varieties are classified on the basis of (1) length of growing season, (2) size and shape of the grain, Genetics and (3) chemical character of the endosperm. Ito Varieties of ordinary cultivated rice, being and Akihama {22) further classified varieties on largely self-pollinated, remain uniform and con- the basis of plant height, straw strength, disease stant if care is taken to preserve pure seed. Muta- resistance, and color of various plant parts. tions, chromosome changes, and natural crossing, On the basis of length of growing season, United however, have brought into existence a wide diver- States varieties grown in the Southern States are sity of types and varieties that affect every part divided into four groups: (1) Very early (100- and function of the plant. Plants may be short or 115 days) ; (2) early (116-130 days) ; (3) midsea- tall enough to grow in deep water ; may have col- son (131-155 days) ; and (4) late (156 days or ored pigments of various shades distributed in more). The length of growing season is somewhat different patterns over the plant ; may tiller little longer in California than in the South. The differ- or profusely ; may have long and slender or short ence in length of growing season between those two and round grains, with or without awns ; and may areas is caused by day length and temperature. have translucent or opaque kernels that differ in The United States varieties are divided into chemical composition. Also, the life cycle of one three grain size and shape classes: short (Pearl), plant may be three times as long as that of another. medium, and long. The more slender long-grain Differences that can be sorted into distinct varieties sometimes are considered a fourth class. contrasting classes make possible the study of seg- Examples of the respective groups are Caloro, regation and recombination as well as linkage rela- Nato, Bluebonnet 50, and Kexoro. (See "Rice tionships of genes, which are the units of heredity. Breeding and Testing Methods in the United For convenience in publishing genetic studies, dif- States,"' p. 22.) ferences that have individual efforts great enough On the basis of chemical characters, rice types to be recognized are assigned gene symbols. To are divided into waxy (glutinous; endosperm con- promote uniform usage among rice geneticists, the tains no amylose) and common (ordinary starchy ; International Rice Commission of the Food and endosperm contains amylose as well as amylopec- Agriculture Organization of the United Nations tin). Only a very small acreage of waxy rice is has adopted and recommended the gene symbols grown in the United States. The percentage of listed below {2). amylose in the starch in the endosperm of varieties Inherited differences, such as plant height, that grown in the United States varies rather widely. are not clear cut but that grade from one extreme ("See Rice Breeding and Testing Methods in the to another involve the interaction of several genes United States," p. 22.) that have similar effects, but the actual number and A more detailed system is needed to classify the the individual contribution of the genes involved many varieties and lines used in breeding studies. remain uncertain. 12 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

Gene Symbols Adopted and Kecommended dw = deep water paddy, so-called floating for Eice ^ E A Ef = Early flowering (low photosensitiv- A, Ad, a = allelic anthocyanin activator genes ity).2See Lf (complementary action with C genes Eu- = Enhancer (intensifier). Precedes produces red or purple in apiculus) symbol of character affected al = albino2 er = erect growth habit, ï-ecessive to An = Awned^ spreading or procumbent. See dw, la au = rudimentary auricle Ex = Exerted vs. enclosed panicle

be = brittle culm Fgr = Fragrant flower Bd = Beaked hull (tip of lemma recurved fs = fine stripe over palea) G Bf = Brown furrow (dark-brown color in g = long glume^ exceeding % length of furrows of lemma and palea). See H^ the spikelet I-Bf = Inhibitor of dark-brown furrow gh = goldhull (golden yellowhull, reces- sive to straw color). See Wh bg = coarse (big) culms Bh = Blackhull (complementary genes)2 gl = glabrous (nonhairy) leaf. See Lh bl = physiologic diseases showing dark- Gm = semi-dominant long glume. More ex- brown or blackish mottled discolora- treme and less regular in expression tion of leaf than g. Epistatic to g bn = bent node—culm forms angle at node green and white stripe; see fs, v. H (H or h is suggested to denote hull, i.e., C, CB, CB^ = allelic basic genes for anthocyanin lerama and palea) CBt, CB^ C color; higher alíeles have pleiotropic H-, Hi, = allelic genes for nonanthocyanin col- expression in internode H«, H^ ors of lemma and palea appearing C alone = tawny-colored apiculus only in the presence of Gh; goldhull CA = red or purple-colored apiculus colors appear with gh CAP = completely and fully purple-colored He = Helminthosporium resistance apiculus hsp = hullspot En-C = Enhancer of C I Ce = Cercospora resistance I = positive vs. negative staining with chl = chlorina (chlorophyll deficiency) iodine-potassium iodide solution Cl = Clustered spikelets, also super cluster I~ = inhibitor (precedes symbol of char- els = cleistogamous spikelets acter inhibited) clw = claw-shaped spikelets. See tri and K (El or k is suggested to denote kernel, Bd i.e., caryopsis) D L (1 (as second letter of a symbol ) is d = dwarfs,2 about % to >^ height of nor- suggested to denote leaf.) mal; discrete classes in segregating 1- =lethal (precedes symbol of character populations having lethal effect) da = double awn la = lazy. See er Dn = Dense or ''compact"; very close ar- Ld = Lodging rangement of spikelets (vs. normal Lf = Late flowering (highly photosensi- panicle); epistatic to Ur. See Lx, tive) .2 See Ef also Cl = liguleless (auricle and collar also Dn2 = Dense vs. lax Ig absent) 2 Dng = Normal vs. lax Dp = Depressed palea and underdeveloped Lh = very hairy (long hair) dominant to palea ordinary pubescence. See gl Ik = grain length (long grain) 2 Imx = extra lemma 1 Recommended by the Eighth (1959) FAO Interna- hi = lutescent tional Rice Commission Working Party on Rice Produc- Lx = Lax vs. normal panicle. See Dn tion and Protection; the list includes present revisions. LX2 = Lax vs. compact 2 More than one gene involved; subscripts to be sup- plied by workers as needed. Letter subscripts are sug- M gested for complementary genes, numeral subscripts for me = multiple embryos (polyembryonic) genes having phenotypically similar effects and also for mp = multiple pistils (polycaryoptic) polymeric genes. mottled leaf; see bl RICE IN THE UNITED STATEIS 13

N tl = twistea leaves nal =narrow leaf tri = triangular hull (spikelet). See clw ni = neckleaf u nk or I-Nk = notched kernel Ur = Undulate rachis vs. normal. See sn V = open hull (parted lemma and palea) V = virescent, also green and white stripe (P (as first letter of a symbol) is sug- Seefs gested to denote anthoeyanin color.) w Exceptions: Ph, Pi wb = white belly (endosperm) = Completely purple apiculus (comple- wb2 = white core (endosperm) mentary action with C and A) Wh = Whitehull, epistatic to gh Pau = Purple auricle, basic to Pig wx = waxy (glutinous) endosperm Pg = Purple outer glumes Y Ph = Phenol staining of hull and bran y = yellow leaf Pi = Pyricularia resistance i-y = lethal yellow, also xantha ^ Pin = Purple internode. See C Z PI = Purple leaf. See Pw z = zebra stripe Pia = Purpleleaf apex (complementary ac- tion with C and A) Pig = Purple ligule. See Pau Plant colors have been studied more extensively I-Pl = colorless leaf except margin than other characters in rice because they are Pn = Purplenode readily classified, and it is possible to analyze pre- Pr = Purplehull (lemma and palea) (com- cisely the genie action and interaction. From the plementary action with CAP) Prp = Purple pericarp economic standpoint, colors are of limited import- Ps = Purple stigma (complementary ac- ance. Eed bran color usually is objectionable as tion with CAP) a mixture in ordinary white rice; purple bran Psh = Purple sheath varieties are used locally for rice wine or special Pu = Purple pulvinus Pw = Purplewash. See PI preparations ; purple leaf varieties have been dis- Px = Purple axil tributed for growing in red-rice infested areas, R so that the weedy type may be identified in the Re = Brown pericarp (basic to Rd) Rd = Red pericarp (complementary action seedling stage and pulled out. with Re) The segregation and interaction of color genes = verticillate (whorled) arrangement of in Japanese varieties have been analyzed thor- rachis C'rinshi'' character) oughly (62), Purple (anthoeyanin) colors, if Rk = Round spikelet (kernel) rl = rolled leaf present, ordinarily are manifest at least in the apiculus. The gradations from dark purple to s = sterility ^ pink and colorless were found to be controlled by Se = Sclerotium oryzae resistance five allelic chromogen genes, C^, C^^, C^\ C^% Se = Photosensitivity. See Ef, also Lf and c. However, C genes alone produce no color Sh = Dominant shattering as in drop-seed or wild types vs. difíicult or inter- in the flowering stage but are responsible for mediate threshing. See th brown colors in decreasing intensity, which ap- Sk = Scented kernel pear as the spikelet matures. sn = sinuous neck. See Ur Activator gene A also must be present to con- spr = spreading panicle branches vert the basic color pigment to purple anthoey- spreading growth habit; see er, also la anin. There are three allels at the A locus. A superclustered spikelets; see Cl (T is suggested to denote plant height third gene, P, present in all but a few varieties, C'Tallness") except where controlled brings out distinct coloration. The numerous by dwarfing genes.) See d colors produced in the apiculus by the inter- I-T = short nondwarf plant vs. tall actions of the C and A allels in the presence of P tb = tipburn yellow are shown in table 4. = Tough dehulling Tf The C^ allel in the presence of A and A^ also th = shattering or easy threshing reces- sive to difficult separation. See Sh has pleiotropic effect, producing color in the inter- 488-871 0—73 2 14 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

TABLE 4.—Apiculus colors resulting from interaction of C and A allels in the presence of the P gene in Japanese rice

Activator Color at time CBP CBt allels CB CBr indicated

fPurple Red purple Red Pink Colorless. __ Flowering. '\ do do do Colorless do Maturity. Ad_ fDeep red Red Orange tinge Orange tinge do Flowering. "\Brown Light brown Yellow tinge Colorless do Maturity. fColorless Colorless Colorless do do Flowering. "\Dark brown Brown Light brown Yellow tinge do Maturity.

Source : Takahashi

node. C allels with A and also less distinctly maturity; and H^ dark furrow color. These colors with A^ result in colored outer glumes. appear only in the presence of the gene for straw- Combinations of C and A are basic to the hull color, Gh ; and with gh, only goldhuU colors appearance of color in other parts of the plant as appear. well as in the apiculus. Purple color of inner Purple bran appears in varieties apparently glumes (lemma and palea), leaves, and nodes was lacking pigmentation in all other parts and seg- shown to result from the complementary action regates in 3:1 ratios in crosses with colorless bran. of C and A with Pr, PI, and Pn, respectively. A cross between a purple bran line that carried a The expected segregation ratio in the Fs of a gene for red bran and a colorless line gave 9 cross between a line with purple apiculus and purple-red:3 purple:3 red:l colorless {9), The inner glumes and a completely colorless line is red underlaid the purple in such a way that it was given below to illustrate the many color patterns possible to separate the first two classes. A cross that appear as the result of recombinations of between a purple bran line having colorless vege- color genes. tative parts and a line having ordinary bran and partly purple leaves segregated in the F2 genera- Ratio of gene Color patterns combinations tion in a ratio of 13 colorless or partly purple 27 C A Pr Purple apiculus and glumes. leaves to three fully purple leaves. The results 9 C A pr Purple apiculus, colorless glumes. show that an inhibitor gene restricts expression 9CaPr Brown (tawny) apiculus and of full purple leaf color. Purple leaf appeared only glumes. in the presence of purple bran. 3 C a pr Brown (tawny) apiculus, colorless Red rice varieties are grown in southern India, in glumes. 16 c All combinations with c are color- Sri Lanka, and in other areas; but red-rice mix- less. tures in white milled rice detract from the appear- ance. Work reported from Japan {^1) showed Many of the segregation ratios previously re- that red bran color required two genes for expres- ported in publications in Japan and other coun- sion. A gene Re for brown bran is basic, and with tries can now be reinterpreted with fewer assump- Rd the bran is bright red. Rd and A are closely linked ; this makes it appear that the apiculus color tions in accordance with the above analysis. Other gene is basic for red. In a cross between a red line workers had recognized two basic complementary of the genetic constitution C A Rd Re and color- color genes, but the pleiotropic action and the less C a rd re, the segregation is 9 red bran, colored existence of multiple allels had not been fully apiculus:3 colorless bran, colored apiculus:3 realized. brown bran, colorless apiculus : 1 both bran and Indian workers {60) have established an allelic apiculus colorless. In this cross, red bran is found series for another set of glume colors: H^, mot- only with colored apiculus and brown bran only tled; HS piebald; H^, green changing to straw at with colorless. In the cross C A Rd Re X c a rd Re, RICE IN THE UNITED STATEiS 15 on the other hand, red bran occurs with and with- F2 segregation was bimodal and indicated the out colored apiculus ; but brown bran never occurs action of one major gene for earliness. with colored apiculus. The ¥2 ratio is 9 red Disease resistance can be analyzed genetically if bran, colored apiculus:3 brown bran, coloriese] distinctly resistant and susceptible classes occur. apiculus : 3 red bran, colorless apiculus : 1 both Specialized races of fungus diseases must be taken bran and apiculus colorless. into account; usually segregation is best deter- The existence of varieties with life cycles that mined following inoculation wdth cultured spores range from as short a period as 3 months to about of individual races. Single and duplicate genes for 10 months contributes greatly to the wide adapta- resistance to four races of Cercospora ory^ae I. bility of rice to difference areas and conditions. Miyake, the fungus that causes narrow brown leaf Varieties are classified for time of flowering as spot, have been determined (27), No linkage was photosensitive or as insensitive. The photosensi- found between genes for Cercospora resistance tive class includes late-maturing varieties, which and the chromosome marker genes C apiculus remain in a vegetative stage while the days are color, H^ furrow color, gh goldhuU, and wx waxy long but probably enter the flowering stage when (26). the night period reaches a critical length. In Sri Several types of genetic segregation have been Lanka, differences in varietal response to varia- reported for inheritance of resistance to Pyri- tions of less than an hour in day length were found Gularia oryzae Cav., the fungus that causes blast. and were controlled by one dominant gene for sen- As early as 1922, resistance was reported to be sitivity (ii ). Early varieties tend to enter the flow- controlled by a single dominant gene {1^0), How- ering stage after the completion of a fairly con- ever, no work has been reported on segregation of stant vegetative stage. resistance to established races. In 1960, a labora- In Japan, a series of six maturity genes domi- tory study of three crosses was reported from nant for lateness and having accumulative effects India (7). Survival was controlled by one or have been postulated (40). The action of these two dominant genes, but the surviving plants and genes is conditioned by temperature. Sampath the parents showed rather high degrees of and Seshu (57) reported an Indian study that susceptibility. dealt with crosses of two varieties from Japan Awns are a conspicuous morphologic character. with photosensitive Indian varieties. One Japa- Although little is known of their physiologic nese variety was insensitive and the other value to the plant, they are supposed to be asso- photosensitive in Japan; but under the tempera- ciated with general hardness {50) and to offer ture and day length conditions at Cuttack, India, some protection against pests. Awns are objec- both flowered early (48 and 68 days from seeding, tionable in threshing and handling the grain. respectively). The Fi plants of crosses of the The development of awns varies considerably Indian varieties with the insensitive Japanese with environmental conditions. Varieties that variety flowered early, and segregation in the F2 show mere tip awns at moderate levels of fertility generation was 3 early : 1 late or 15 early : 1 late. may develop much more prominent awns at However, in crosses between the photosensitive higher levels. In Louisiana, the Caloro variety Japanese variety and Indian varieties the Fi develops awns when seeded very early but may be plants flowered late and segregation in the F2 practically awnless when seeded late. In the F2 generation was 1 early : 3 late or 1 early : 15 late. generation, ratios of awned to awnless plants of It was assumed that in the crosses with the photo- 3 :1,15:1, and 9: 7 have been reported from India, sensitive parent, genes involved in temperature Japan, and the United States. These ratios show, responses and modifying the photoperiodic re- respectively, that single, duplicate, and comple- sponse were segregating. These authors cite a mentary genes act to produce awns. A study in Japanese study in which a gene for late flowering California {28) showed that fully awned plants was found to be epistatic to one for early differed from awnless by two genes, fully awned flowering. from partly awned by one gene, and partly awned In Louisiana, a cross betv/een varieties that from awnless by one gene. headed in 90 and 125 days was studied (8). The IMost rice varieties have pubescent leaves and 16 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

hulls, although a few are essentially glabrous. erous F3 progenies, each comparatively uniform Pubescene has been eliminated from the vari- in number of tillers, led to the same conclusion. eties grown in the . In Standing ability or lodging resistance is a com- most varieties, a single gene pair Gl : gl controls plex character that is difficult to measure in in- the development of plant hairs on all outer sur- heritance studies. In a study reported from faces of the plant. However, some varieties have India (50), the F2 progenies of a cross segregated some pubescence on the leaf surface, although the 3 lodging:! nonlodging, but the latter were low lemma and palea are smooth. Nagao, Takahashi, tillering and late maturing. and Kinoshita (^) reported the segregation in In Louisiana, a cross between an early-matur- crosses of the latter type with ordinary pubescent ing, medium-grain selection and the late-maturing Japanese types. Three classes of pubescene plus variety Rexoro indicated transgressive segrega- glabrous appeared in the F2 generation in ratios tion for yield (S). F4 lines were recovered that of 9: 3: 3:1 and 27: 9: 21: 3. It was concluded that had yields significantly higher than those of the two genes Hla and Hlb together but not singly high-yielding parent, but none had yields sig- bring about the development of hairs on the leaf nificantly lower than those of the low-yielding surfaces even in the presence of gl. parent. Higher yield was strongly associated The strength of attachment of spikelets to their with earlier maturity. No correlation was found pedicles is of practical importance to the grower. between yield and spikelet length and breadth, or Shattering (shedding) of grain in the wind, between yield and grain weight. which occurs in red rice, would make it impos- In correlation studies from other countries sible to harvest the crop; whereas at the other Ramiah and Rao (SO) indicate association of extreme, which occurs in some of the Japanese liigher yield with number of tillers, number of varieties, the attachment is so tight that in com- grains per ear, and plant height. bining much grain would remain on the straw. Genetic information on the quality character of Types that thresh free of the straw yet do not rice is very limited, although the single gene seg- shatter easily are needed for mechanized harvest. regation between common and waxy types was Easy threshing Sh is dominant to intermediate one of the first characters studied. The inherit- or tough threshing sh; but, on the other hand, ance of amylose content as indicated by the iodine tough threshing is dominant to intermediate value was studied in a cross of Toro (high iodine threshing. value) X Texas Patna (low iodine value) (58), Multiple gene inheritance of grain length was The value in the Fi generation was low, indicat- reported in a study in the United States (£9), In ing partial dominance. A bimodal distribution crosses of short X medium, short X long, and was obtained in the F2, indicating dominance of medium X long types, variation in length in the one major gene for low iodine value. An asso- F2 generation was continuous. Similar results, ciation was shown between low iodine value and colored apiculus ; the latter is also a character of also showing transgressive segregation, were - the low iodine parent. Recovered lines with high tained for breadth of grain. iodine value were much more numerous in the Tillering ability is of great importance where early-maturing than in the late-maturing group. rice is transplanted, but much less so where direct Linkage refers to the closeness of association in seeding at relatively high rates is practiced. inheritance of genes located on the same chro- Ramiah and Rao (SO) reported from India that mosomes. Genes controlling characters that seg- the Fi plants of three crosses were intermediate regate into clear-cut classes and that are viable for number of tillers per plant, and the F2 prog- are useful for determining linkage relationships. enies showed transgressive segregation. In one Several of the genes for color are suitable marker cross high correlation between the number of genes. Easily classified morphologic characters tillers in F2 plants and their respective F3 prog- dependent on single genes are also useful, includ- enies indicated that the controlling genes are ing some of the more conspicuous characters such limited in number—probably no more than three as awns, long outer glumes (of which recessive or four. In another cross the occurrence of num- and semidominant types are known), pubescence. RICE IN THE UNITED STATEB 17 neckleaf, liguleless, lazy (ageotropic), brittle from other studies and reviews (^, 23^ 2If.^ 25^ 38^ culm, numerous dwarfs, waxy endosperm, and 63) are provisionally included with the groups to viable types of chlorophyll deficiencies. which they appear to belong. Linkages between In a report from Japan, Nagao and Takahashi genes that have been reported but not definitely (4^) presented data that established 12 linkage located on the respective chromosomes are listed groups, representing, as expected, the number of in table 5. Linkages between genes that have chromosomes found in rice. Takahashi {63) has been reported but have not been assigned to a since amplified the information on the linkage linkage group are also shown in table 5. Link- groups. The linkage diagrams (maps) shown in age maps of rice are much less complete than figure 5 are based on the Japanese results. Data those of barley and maize.

46

38 40 ■V 41 22 ,' 23 25 35 WAXY CI

52 .A- 47 28 _A_ 38 49 _A_ ^r -TV. 25 35 31 7 , 24 PURPLE LEAF PI Ig Ph Pr

30

;^°^27~^, 39 ACTIVATOR i LAZY I- A Rd Pn ia th

30 / ' V 34 NECKLEAF LONG GLUME I '^ I I if> g Re

25 42 BLACKLEAF INHIBITOR bl ds I-Bf 40 48 -VU / 18 , 28 44 1^ DWARF BRITTLE be dgAni dl gh 46 -A- 4] 28 41 GLABROUS FINE STRIPE X An fs Dn gl riGUEE 5.—Rice linkage groups. 18 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

TABLE 5.—Linkages not definitely located on the respective chromosomes

Linkages tentatively Percentage Linkages tentatively Percentage assigned to groups recombina- Authority assigned to groups recombina- Authority tion 1 tion 1

WAXY GROUP LONG GLUME GROUP wx VS. Pia 45 Nagao and Taka- Con. hashi U2). Rk vs. I-An 31? Jodon (unpub- C vs. H^ 29 Ramiah and Rao lished) . Pin vs. Re Unknown... Ramiah and Rao C vs. Se 16 Chandraratna (IQ). (50). C vs. Lf 32 Yamaguchi (^5). Revs. LfJ 22 Yamaguchi (^á»). C vs. dp Unknown __ Nagamatus and INHIBITOR GROUP Ohmura.i Psvs. Psh 10 Chao (^^). C vs. n do Do.^ Ps vs. Apic 9 Shafi and Aziz (5P). 1 vs. d2i 17 Jodon (unpub- Apic vs. Psh 12 Do. lished) . Ps vs. Pr 2 D'Cruz (Í7). wx vs. Xi Unknown __ Oka (1953).i wx vs. yi do Do.i DWARF GROUP PURPLE LEAF GROUP gh VS. Ana 43 Nagao and Taka- PI VS. An 29 Jodon (unpub- hashi (42). lished) . gh vs. P or Pr 30 Jodon and Chilton PI vs. lop Unknown __ Nagamatsu and (26). Ohmura, 1961.i FINE STRIPE GROUP Ig vs. Wh 8 Jodon (24, ^5). None Ig vs. Xa Unknown.-- Nishimura.i LAZY GROUP Pr vs. P 32 Jodon (unpub- La vs. d Unknown... Nagamatsu and lished) . P vs. (blackspot) 33 Do. Ohmura.i Pr vs. Ig 30 Richhariaand NECKLEAF GROUP others (51). nl VS. An 32? Jodon (unpub- Prvs. Px 9 Do. lished) . Pxvs. Psh 6 Do. GLABROUS GROUP Prvs. gh 32 Jodon (24, 25). gl vs. H« 39 Nagao and Taka- Pr vs. h 36 Do. hashi (42). gh vs. Wh Unknown... Do. LINKAGES NOT Ps vs. Pr do Hsieh (20). ASSIGNED TO d vs. PI do Do. LINKAGE GROUP Pi vs. Ig do Do. Rk VS. An 14 and 24... Anandan (6). ACTIVATOR GROUP Bhvs. An 21 Kuang (5^). Avs. bn 23 Hsieh (iP). Bh vs. Pin? 17 Parnell, Ranga- bn vs. d22 28 Do. swami Ayyangar, d22 vs. Igt 16 Do. and Ramiah (47). Igtvs. bn 30 Do. An vs. d 11 Kuang (36). A vs. d22 37 Do. Pr?vs. Pig? 13 Hebert (i^). Rd vs. Pin 21 Ramiah and Rao Pr?vs. Pin 10 Do. (50). Purple collar vs. Prp.. 13 Breaux (9). LONG GLUME GROUP Psh vs. Prp 16 ComesLMX (16). dfl VS. d: 39 Nagao and Taka- bl vs. Prp Unknown... Jodon (unpub- hashi (42). lished) . g VS. Rk 8 Kadam and D'Cruz d2i vs. Prp do Do. (31). sk vs. Prp do Ramiah and Rao Apic vs. g 17 Do. (50).

1 As reported by Takahashi RICE IN THE UNITED STATEiS 19

Selected References (15) CHEVALIER, A. 1932. NOUVELLE CONTRIBUTION A L'ETUDE SYSTE- (1) ANONYMOUS. MATIQUE DES ORYZA. Rev. de Bot. Appl. et 1950. WORLD CATALOG OF GENETIC STOCKS—^RICE. d'Agr. Trop. 12: 1014-1032. Food and Agr. Organ. United Nations, 30 (16) COMEAUX, D. J. pp. [Mimeographed.] 1946. AN INHERITANCE AND LINKAGE STUDY OF (2) VIRESCENCE AND OTHER FACTORS IN RICE, 1963. RICE GENE SYMBOLIZATION AND LINKAGE ORYZA SATIVA. [Unpublished M.S. thesis, La. GROUPS. U.S. Dept. Agr. Agr. Res. Serv. State Univ., 50 pp.] ARS 34-28, 56 pp. (17) D'CRUZ, R. (3) AD AIR, 0. R. 1960. A LINKAGE BETWEEN TWO BASIC GENES FOR 1934. STUDIES ON BLOOMING IN RICE. AgrOn. ANTHOCYANIN COLOUR IN RICE. Sci. and Jour. 26: 965-973. Cult. [India] 25: 534-536. (4) . (18) HEBERT, L. P. 1936. STUDIES ON GROWTH IN RICE. Agron. Jour. 1938. A GENETIC STUDY OF COLOR AND CERTAIN 28: 506-514. OTHER CHARACTERS IN RICE. [Unpublished (5) AKIMOTO, S., and TOGARI, Y. M.S. thesis, La. State Univ., 57 pp.] 1939 VARIETAL DIFFERENCES IN PANICLE DEVELOP- (19) HSIEH, S. C. MENT OF RICE WITH REFERENCE TO EARLY OR 1960. GENIC ANALYSIS IN RICE. I. COLORATION GENES LATE TRANSPLANTING. Crop Sci. Soc. Japan, AND OTHER CHARACTERS IN RICE. Acad. Siuica Proc. 11(1) : 7-14. Bot. Bul. 1:117-132. (6) ANANDAN, M. (20) 1928-33. RPTS. ON AGR. STAS., MADRAS DEPT. AGR., 1961. ANALYSIS OF GENES FOR BLAST DISEASE RE- ADUTURIA, 1927-28 AND 1932-33. [Original SISTANCE CAUSED BY PIRICULARIA ORYZAE. not see. Reported by Ramiah and Rao Sympos., studies on Cause of Low Yield of (SO).] Rice in Tropical and Sub-Tropical , (7) BHAPKAR, D. G., and D'CRUZ, R. Proc. Spec. Pub. Taiwan Agr. Res. Inst. 3 : 1960. INHERITANCE OF BLAST RESISTANCE IN RICE. 45-52. Poona Agr. Col. Mag. 51(2) : 23-25. (21) (8) BOLLICH, 0. N. HuTCHiNSON, J. B., RAMIAH, K., and MEMBERS OF Two SPECIAL SUBCOMMITTEES. 1957. INHERITANCE OF SEVERAL ECONOMIC QUANTI- 1938. THE DESCRIPTION OF CROP-PLANT CHARACTERS TATIVE CHARACTERS IN RICE. [Unpublished AND THEIR RANGES OF VARIATION. INTRODUC- Ph.D. thesis. La. State Univ., 129 pp.] TION. Indian Jour. Agri. Sei. 8(5) : 567-569; (9) BREAUX, NORRIS. IL VARIABILITY IN RICE. 8(5) : 592-616. 1942. INDEPENDENT ASSORTMENT, INTERACTION OF (22) ITO, H., AKIHAMA, T. FACTORS AND LINKAGE STUDIES IN THE FA and OF RICE CROSS. Proc. La. Acad. Sei. 6: 52- 1962. AN APPROACH FOR THE SYMBOLIZATION OF 59. Feb. 15,1942. COLORS IN RICE PLANTS AND ITS ADOPTION FOR (10) CHANDRARATNA, M. F. THE CLASSIFICATION OF RICE VARIETIES. Jap. 1953. A GENE FOR PHOTOPERIOD SENSITIVITY IN RICE Jour. Breeding 12: 221-225. [In Japanese. LINKED WITH APicuLUS COLOUR. Nature English summary.] 171: 1162-1163. (23) JODON, N. E. (11) • 1948. SUMMARY OF RICE LINKAGE DATA. U.S. Dept. 1955. GENETICS OF PHOTOPERIODIC SENSITIVITY IN Agr., Bur. Plant Indus., Soils, and Agr. RICE. Jour. Genet. 53: 215-223. Engin., 34 pp. [Mimeographed.] (12) CHAO, L. F. (24) 1928. LINKAGE STUDIES IN RICE. Geuetics 13 : 1955. PRESENT STATUS OF RICE GENETICS. Agr. 133-169. Assoc. China, Jour, (n.s.) 10: 5-21. (13) CHATTERJEE, D. (25) 1948. A MODIFIED KEY AND ENUMERATION OF THE 1956. PRESENT STATUS OF RICE GENETICS. Agr. SPECIES OF ORYZA LINN. Indian Jour. Agr. Assoc. China, Jour, (n.s.) 14: 69-73. Sei. 18(3) : 185-192. (26) and CHILTON, S. J. P. (14) 1946. SOME CHARACTERS INHERITED INDEPENDENTLY 1951. NOTE ON THE ORIGIN AND DISTRIBUTION OF OF REACTION TO PHYSIOLOGIC RACES OF CERCO- WILD AND CULTIVATED RICES. Indian Jour. SPORA ORYZAE IN RICE. Amer. Soc. Agron. Genet, and Plant Breeding 11(1) : 18^22. Jour. 38: 864-872. 20 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

(27) ■ RYKER, T. C, and CHILTON, S. J. P. (42) and TAKAHASHI, M. 1944. INHERITANCE OF REACTION TO PHYSIOLOGIC 1963. TRIAL CONSTRUCTION OF TWELVE LINKAGE RACES OF CERCOSPORA ORYZAE IN RICE. Amer. GROUPS IN JAPANESE RICE. Hokkaido Imp. Soc. Agron. Jour. 36: 497-507. Univ., Faculty Agr. Jour. 53: 72-130. (28) JONES, J.W. (43) TAKAHASHI, M., and KINOSHITA, T. 1933. INHERITANCE OF CHARACTERS IN RICE. Jour. 1960. GENETICAL STUDIES ON RICE PLANT. XXV. Agr. Res. 47 : 771-782. INHERITANCE OF THREE MORPHOLOGICAL (29) ADAIR, C. R., BEACHELL, H. M., and DAVIS, L. L. CHARACTERS. PUBESCENCE OF LEAVES AND 1935. INHERITANCE OF EARLINESS AND LENGTH OF FLORAL GLUMES, AND DEFORMATION OF KERNEL IN RICE. Amer. Soc. Agron. Jour. 27 : EMPTY GLUMES Hokkaido Imp. Univ., 910-921. Faculty Agr. Jour. 51: 299-314. [In Eng- (30) JULIANO, J. B., and ALDAMA, M. J. lish.] 1937. MORPHOLOGY OF ORYZA SATIVA LINNAEUS. (44) NANDI, H. K. Philippine Agr. 26: 1-134. 1936. THE CHROMOSOME MORPHOLOGY, SECONDARY (31) KADAM, B. S., and D'CRUZ, R. ASSOCIATION AND ORIGIN OF CULTIVATED RICE. 1960. GENIC ANALYSIS IN RICE. in. INHERITANCE OF Jour. Genet. 33(2) : 315-336. SOME CHARACTERS IN TWO CLUSTERED VARIE- (45) OKA, H. I. TIES OF RICE. Indian Jour. Genet, and Plant 1958. INTERVARIETAL VARIATION AND CLASSIFICA- Breeding 20: 79-84. TION OF CULTIVATED RICE. Indian Jour. ( 32 ) KATAYAMA, TUKUDA. Genet, and Plant Breeding 18 (2) : 79-89. 1931. ANALYTICAL STUDIES OF TILLERING IN PADDY RICE. Jour. Imp. Agr. Exp. Sta. 1(4) : 327- (46) and CHANG, W. T. 1962. RICE VARIETIES INTERMEDIATE BETWEEN WILD 374. [English summary, 371-374.] AND CULTIVATED FORMS AND THE ORIGIN OF (33) KATO, S., KOSAKA, H., and HARA, S. THE JAPÓNICA TYPE. Acad. Siuica Bot. 1928. ON THE AFFINITY OF RICE VARIETIES AS SHOWN BY THE FERTILITY OF HYBRID PLANTS. Bul., 3(1) : 109-131. Kyushu Imp. Univ., Fakult. Terkult. Bui. (47) PARNELL, F. R., RANGASWAMI AYYANGAR, G. N., Sei. Rpt. 3: 132. [In Japanese.] and RAMIAH, K. (34) KOSAKA, H., and HARA, S., and others. 1917. THE INHERITANCE OF CHARACTERS IN RICE. 1930. ON THE AFFINITY OF THE CULTIVATED VARIE- L India Dept. Agr. Mem., Bot. Ser. 9 : TIES OF RICE PLANTS, ORYZA SATIVA L. KyUSllU 75-105. Imp. Univ., Dept. Agr. Jour. 2(9) : 241-276. (48) RAMIAH, K., and GHOSE, R. L. M. (35) KlHARA, H. 1951. ORIGIN AND DISTRIBUTION OF CULTIVATED 1959. CONSIDERATIONS ON THE ORIGIN OF CULTI- PLANTS OF SOUTH ASIA—^RicE. Indian Jour. VATED RICE. Seiken Ziho. 10: 68-83. Genet, and Plant Breeding 11(1) : 7-13. (36) KuANG, H. H. (49) and NARASIMHAN, M. 1951. STUDIES ON RICE CYTOLOGY AND GENETICS AS 1936. DEVELOPMENTAL STUDIES IN RICE. I. Madras WELL AS BREEDING WORK IN CHINA. AgrOU. Agr. Jour. 24(2) : 50-66. Jour. 43: 387-397. (50) and RAO, M. B. V. N. (37) MATSUDA, K. 1953. RICE BREEDING AND GENETICS. Indian Council 1929. ON THE DEVELOPMENT OF RICE KERNELS. Agr. Res. Sei. Monog. 19, 360 pp. Nogaku Kwaiho (Sei. Agr. Soc. [Japan] (51) RICHHARIA, R. H., MISRO, B., BUTANY, W. T., and Jour.), No. 314, 34 pp. SEETHARAMAN, R. (38) MATSUURA, H. 1960. LINKAGE STUDIES IN RICE (ORYZA SATIVA L.). 1933. ORYZA SATIVA. Hi8 A Bibliographical Mon- Euphytica 9: 122-126. ograph on Plant Genetics (Genie Analy- (52) RODRIGO, P. A. sis), 1900-1929. Ed. 2, pp. 240-265. Hok- 1925. POLLINATION AND THE FLOWER OF RICE. kaido Imp. Univ., Sapporo. Philippine Agr. 14(3) : 155-171. (39) MiZUSHIMA, U. (53) RoscHEvicz, R. J. 1948. STUDY ON SEXUAL AFFINITY AMONG RICE 1931. A CONTRIBUTION TO THE KNOWLEDGE OF RICE. VARIETIES, ORYZA SATIVA L. I. ANALYSIS OF AFFINITY OF JAPANESE, AMERICAN, AND JAV- Bul. Appl. Bot., Genet., and Plant Breed- ANESE VARIETIES. Seibutu 3(2) : 41-52. ing 27(4) : 1-133. [English summary, pp. (40) NAGAI, I. 119-133.] 1959. JAPÓNICA RICE. ITS BREEDING AND CULTURE. (54) SAKAI, K. I. 843 pp. Yokendo, Ltd., Tokyo. 1935. CHROMOSOME STUDIES IN ORYZA SATIVA L. (41) NAGAO, S. I. THE SECONDARY ASSOCIATION OF THE 1951. GENIC ANALYSIS AND LINKAGE RELATION- MEIOTIC CHROMOSOMES. Jap. Jour. Genet. SHIP OF CHARACTERS IN RICE. In Demcrec, 11(3) : 145-156. [English summary, p. M., ed.. Adv. in Genet. 4: 181-212. 154.] RICE IN THE UNITED STATEIS 21

(55) SAMPATH, S., and GOVINDASWAMI, S. (64) TATEOKA, T. 1958. WILD RICES OF ORISSA—THEIR RELATIONSHIP 1963. TAXONOMIC STUDIES OF ORYZA. III. KEY TO TO CULTIVATED VARIETIES. Rice News Teller, THE SPECIES AND THEIR ENUMERATION. Bot. July. Mag. Tokyo 76: 165-173. (56) and RAO, M. B. V. N. (65) TERADA, S. 1951. INTERRELATIONSHIPS BETWEEN SPECIES IN 1928. EMBRYOLOGICAL STUDIES IN ORYZA SATIVA L. THE GENUS ORYZA. Indian Jour. Genet, and Hokkaido Imp. Univ., Col. Agr. Jour. Plant Breeding 11: 14-17. 19(4) : 245-260. [In English.] (57) and SESHU, D. V. (66) TERAO, H., and MIDUSIMA, U. 1961. GENETICS OF PHOTOPERIOD RESPONSE IN RICE. 1939. SOME CONSIDERATIONS ON THE CLASSIFICA- Indian Jour. Genet, and Plant Breeding TION OF ORYZA SATIVA L. INTO TWO SUB- 21 : 38-42. SPECIES SO-CALLED 'JAPÓNICA' AND 'INDICA.' (58) SEETHARAMAN, RAMASWAMY. 1959. THE INHERITANCE OF IODINE VALUE IN RICE Jap. Jour. Bot. 10(3) : 213-258. AND ITS ASSOCIATION WITH OTHER CHARAC- (67) TING, YING. TERS. [Ph.D. Diss., La. State Univ., 65 pp.] 1949. ORIGINATION OF THE RICE CULTIVATION IN (59) SHAFI, M., and Aziz, M. A. CHINA. Rice Expt. Sta., Sun Yatsen Univ., 1959. THE INHERITANCE OF ANTHOCYANIN PIG- Canton, China. Col. of Agr., Agron, Bui. 7, MENT IN THE OUTERGLUME AND APlCULUS Pub. series III, 18 pp. [Resume in Eng- OF RICE. Agr. Pakistan 10: 217-232. lish]. (60) SHASTRY, S. V. S., RAO, D. R. R., and MISRA, R. N. 1960. PACHYTENE ANALYSIS IN ORYZA. I. CHRO- (68) YAMAGUCHI, Y. MOSOME MORPHOLOGY IN ORYZA SATIVA. In- 1921. ETUDES D'HéRéDITé SUR LA COULEUR DES dian Jour. Genet, and Plant Breeding GLUMES CHEZ LE RIZ. Bot. Mag. [Tokyo] 20(1) : 15-21. 35: 106-112. [English review in Internatl. (61) SMITH, W. D., DEFFES, J. J., BENNETT, C. H., and Rev. Sei. and Pract. Agr. [Rome] 12: others. 1399-1401.] 1938. EFFECT OF DATE OF HARVEST ON YIELD AND (69) MILLING QUALITY OF RICE. U.S. Dept. Agr. 1929. FURTHER CONTRIBUTIONS TO THE KNOW^L- Cir. 484, 20 pp. EDGE OF THE SECOND (S.M.) LINKAGE GROUP (62) TAKAHASHI, M. IN RICE. Nôgaku Kenkyu [Studies in Agr. 1957. ANALYSIS ON APlCULUS COLOR GENES ESSEN- Sei.] Japan 13: 135-172. [In Japanese.] TIAL TO ANTHOCYANIN COLORATION IN RICE. Hokkaido Imp. Univ., Faculty Agr. Jour. (70) YEH, B., and HENDERSON, M. T. 50: 266-362, 6 plates. [In English.] 1961. CYTOGENETIC RELATIONSHIPS BETWEEN CUL- (63) TIVATED RICE, ORYZA SATIVA L., AND FIVE 1964. LINKAGE GROUPS AND GENE SCHEMES OF SOME WILD SPECIES OF ORYZA. Crop Sci. 1: 445- STRIKING MORPHOLOGICAL CHARACTERS IN 450. JAPANESE RICE. In Rice Genetics and Cy- YUNG, C. T. togenetics, Sympos. Proc, Internatl. Rice (71) 1938. DEVELOPMENTAL ANATOMY OF THE SEEDLING Res. Inst., Los Banos, Philippines, pp. OF THE RICE PLANT. Bot. Gaz. 99(4) : 215-236. Elsevier Pub. Co., Amsterdam, London, and New York. 786-802. RICE BREEDING AND TESTING METHODS IN THE UNITED STATES

By C. ROY ADAIB, C. N. BOLLICH, D. H. BOWMAN, NELSON E. JOSON, T. H. JOHNSTON, B. D. WEBB, and J. G. ATKINS

History and Objectives Stoneville, Miss., about 1951. In 1969, a new rice Eice improvement in the United States has been research facility was established at the Univer- discussed by Jones (69), and the descriptions and sity of California at Davis, Calif. Rice-breeding performance of varieties developed have been investigations are conducted cooperatively by the reported by several authors (/, ^, ^7, 28, 82, ^5, U.S. Department of Agriculture and the State 49,51, 57, 61, 65, 71, 73, 82). Although some work agricultural experiment station at each of these to improve United States rice varieties was done locations. The rice-breeding studies were started before 1909, it was not until about then that com- in Louisiana in 1909, in California in 1912, and in prehensive, cooperative rice-breeding studies were Arkansas in 1931. Although a comprehensive rice- started in the United States. Objectives of the breeding program was not started in Texas until rice-breeding program and methods used have 1931 and in Mississippi until 1958, some breeding been described ( 9, 20, 69, 84, 88). According to a work had been done in these States before these report by the Rice Millers Association, all rice dates. A rice genetic and physiology program was varieties identified in the acreage report as grown initiated at Davis, Calif., in 1969. in the United States in 1970 evolved from these Most of the early work consisted of testing cooperative breeding experiments. selections from foreign introductions and com- Although introductions were made by individ- mercial fields. Many varieties, such as Caloro, uals after the original introductions into South Colusa, Fortuna, Nira, and Rexoro, were devel- Carolina in the 17th century, little eifort was made oped and released from 1909 to about 1937. by Federal or State agencies to improve rice vari- S. L. Wright, a farmer in Louisiana, developed eties until work was started by the U.S. Depart- several rice varieties by selection. These varieties ment of Agriculture in 1899. At that time Seaman probably were progenies of natural hybrids be- A. Knapp, an explorer in the Division of Botany, tween varieties, such as the long-grain Honduras introduced from Japan 10 tons of Kyushu rice that and the short-grain Shinriki. The most widely was distributed in southwestern Louisiana and grown varieties developed by Wright were Blue probably in eastern Texas, where he arranged Rose and Early Prolific, which are medium- farm demonstrations of rice varieties and cultural grain types, and Edith and Lady Wright, which methods. Many varieties were introduced and are long-grain types. None of these varieties tested by Department workers on demonstration were grown in 1963, but they were the leading farms in Louisiana and Texas, and later in Arkan- varieties in the southern rice area from about sas and California, before the establishment of 1915 to 1940. rice experiment stations in these States. During the early years, no attempt was made The Rice Experimental Station at Crowley, La., to improve varieties by hybridization. Although and the Rice Experiment Station at Beaumont, a few hybrids had been made at experiment sta- Tex., (now the Texas A. & M. University, Agri- tions in the southern rice area at an earlier date, cultural Research and Extension Center) were es- it was not until after 1922 that this method of tablished in 1909. The Biggs Rice Field Station breeding was used. at Biggs, Calif., (now the Rice Experiment Sta- The primary objective in rice breeding in the tion of the California Cooperative Rice Research United States is to develop varieties that will Foundation, Inc., an industry-supported agency) assure a maximum and stable production of the was established in 1926. Rice investigations were types of rice required by producers and consumers. started at the Delta Branch Experiment Station, Emphasis is given to developing shor't-season 22 RICE IN THE UNITED STATE.S 23 varieties. Short-, medium-, and long grain types— Cross levees at designated points divide the field with a wide maturity range within each grain into areas of approximately one-tenth acre that type class—should be developed. The objectives of can be irrigated or drained independently (fig. 6). that program are to develop varieties that (1) Larger blocks can be formed by eliminating one germinate quickly and grow rapidly in the seed- or more cross levees. Irrigation laterals parallel ling stage; (2) tolerate low temperature in the roadways and are located midway between the germinating, seedling, and flowering stages and roadways or directly adjacent to one side of the tolerate low irrigation water temperatures during roadways. the entire growing season; (3) are resistant to Kicefield machinery is used to build levees. In alkaline and saline soils and to salt in the irriga- heavy clay soils, a steel-bladed levee builder (fig. tion water; (4) are resistant to diseases and in- 7^ Ä) is used; in the more loamy soils, a disk-type sects; (5) have short, stiff straw and resist levee builder (fig. 7, 5) is used. Usually the cross lodging; (6) respond to and make efficient use of levees have an 8- to 10-foot base. A concrete levee maximum rates of fertilizer; (7) mature uni- packer is useful in packing newly constructed formly and produce seed that has a period of levees. A concrete tile, 4 to 6 inches in diameter, is dormancy so that the grain will not germinate placed at the lower end of each block to drain the when harvest is delayed by rain; (8) produce water into the road ditch so that it will hold the maximum field and milling yields when grown irrigation water. The tile is closed with a metal under a wide range of environmental conditions ; plate. Sometimes plastic tube siphons are used to (9) have the desired cooking and processing quali- irrigate from the lateral or to drain the water into ties required by domestic and foreign trade; and the road ditch. In other cases, plastic pipe 4 inches (10) have maximum levels of good-quality protein. in diameter is inserted into the levee to irrigate Each of these objectives is important in each from the lateral (fig. 8), and plastic pipe 2 inches rice-producing area in the United States, although in diameter is used to drain the water into the road some of the objectives are more important in one ditch. The pipes are closed with rubber stoppers area than in another. For example, cold toler- or plastic bag covers held in place with rubber ance, especially tolerance to cold irrigation water, bands or ties. is more important in California than in the south- In the southern rice area, it is customary to ern rice area. Disease resistance, on the other hand, plow down the cross levees after each rice crop is more important in the southern rice area than and to cultivate the fields with ricefield equip- in California. ment. The irrigation laterals and border levees may or may not be torn down. After cross levees have been plowed down, the areas are disked or Cultural Methods and Equipment for plowed and are leveled with a land plane. Breeding Rice in the United States A 3-year rotation system is practiced in Land for rice-breeding exxperiments should be Arkansas and Louisiana. In Arkansas, the land of uniform soil type and topography. A gradual may be cropped to soybeans the first year. The soy- and uniform slope is desirable to afford reason- beans may be harvested or turned under or disked ably rapid drainage when necessary. However, into the ground as green manure, after which a the slope should be such that blocks of three- winter grain crop may be sown. The grain crop tenth acre or larger can be uniformly irrigated may be overplanted with lespedeza the following without contour levees. spring. The grain crop is harvested in early sum- In the United States it is customery to locate mer, and the lespedeza is harvested for hay or roadways at 200- to 300-foot intervals running in seed in the summer or fall. The field is plowed or the direction of the fall of the land and wide disked, and land is leveled during the winter and enough to accommodate field equipment and other early spiking preparatory to seeding rice the fol- vehicles. The road ditches are used for drainage lowing year. In Louisiana, the land is fallowed and are paralleled by border levees. The entire when not in rice. area should be graded to a uniform slope before A 3-year rotation system is also practiced in roadways and irrigation laterals are constructed. Texas. It consists of 1 year of rice followed by 2 24 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

'^ß0:m^i^Mi*m^i^^M,■■^aÄp

'•^^^^^MSfá

FiGUEE 6.—Seeded nursery plots. Some plots are irrigated.

years of fallow. Land is leveled during the second Permanent levees are maintained in some of the summer as time and weather permit. areas used for rice nurseries. In other areas, On clay soils the cross levees are reconstructed plastic covered levees are used, or earthen levees in the late fall or winter before the fields are are constructed with a bulldozer or front-end load- seeded again with rice. After cross levees are con- ing tractor each year before seeding. structed, the land is prepared by using lift-type Weeds and grass are controlled on lateral and implements attached to a lightweight farm tractor. drainage ditch banks, levees, and alleyways in rice On silt loam soil the cross levees may be made by spraying with a contact herbicide. immediately after seeding. Various types of seeding equipment are used. In California, a 2-year rotation generally is used In field plot varietal experiments, seeds are sown in rice-breeding nurseries. The land is plowed in with small grain drills or specially equipped the winter or early spring after the rice crop has drills. In yield trials on nursery plots or in been grown and harvested. The fields usually are experiments where multiple seed sources are in- irrigated one or more times during the summer volved, belt-type seeders (fig. 9) or other readily cleaned drills are used for seeding. Breeding when they are not cropped to rice to germinate nurseries involving numerous panicle-row selec- weed, grass, and rice seeds. The field is cultivated tions are sown by dribbling seed into a single- after each irrigation to destroy the plants that row manually operated drill (fig. 10) or a multi- emerge. ple-row power-operated drill. RICE IN THE UNITED STATES 25

On experimental plots uniform application of When dried by heated air, the bundles are placed known amounts of fertilizer materials is difficult. in drying trays as harvested and dried over a tun- A basal application of phosphorus and potassium nel or rack-type drying unit with forced heated may be applied with a conventional fertilizer dis- air (fig. 15). Bundles are left in the trays until tributor before seeding rice or to one of the other the grain is ready to be threshed. crops grown in rotation with rice. Although the fertilizer usually is distributed by a tractor- drawn implement, it may be distributed by air- plane. In the southern rice area, the nitrogen fertilizer for breeding trials usually is applied as ^i» a topdressing, although sometimes part or all is applied at seeding time. Various methods and devices have been used to apply fertilizer to rice grown in breeding nurseries. Weighed amounts of fertilizer are applied by hand as a topdressing. Mechanized fertilizer distributors have been tried with only limited success. Most devices are slower than hand application, do not apply or distribute uni- form amounts of fertilizer, and can be used only on relatively dry soil. At Stuttgart, Ark., a method of topdressing with a liquid solution has been developed for nursery areas that are reasonably level but have a very gentle slope. A small nursery area is drained overnight and the nitrogen solution is metered into the water after the borrow ditches have been filled (figs. 11 and 12). As soon as the measured amount of solution has completed drip- ping, the watering pipe is closed off for at least 24 hours. A manually operated "planter" is used to apply nitrogen fertilizer between alternate rows when the soil is dry (fig. 13). For a number of years a fertilizer distributor attached to a 3-row seeder was used at Beaumont, Tex. This drill used regular grain drill fertilizer distributor parts. Top-feed and other hopper- type distributors have been used with limited success to topdress plots. A force-feed-type dis- tributor that drops the fertilizer on the soil surface is fairly successful (ñg. 14). Cross-plot fertilizing with this type distributor appears to be desirable in overcoming irregularities of distribution. Because rice nurseries are likely to be muddy at harvesttime, mechanical harvesting equipment is not used. Hand sickles are used to harvest small plots ; combine harvesters may be used to harvest large plots. Bundles from small plots are air dried by hang- FIGURE 7.—Constructing levees in nursery with (A) steel- ing in a drying shed or by using heated-air driers. bladed and (ß) disk-type levee builder. 26 AGRICULTURE HANDBOOK NO. 2 8 9, U.S. DEPT. OF AGRICULTURE

S*^ '^^li^f more important than temperature in retaining via- bility. Eice with a moisture content of 10 percent or below (wet basis) retained viability for several years at Beaumont, Tex. The temperature in the storage room ranged from 70° to 90° F., and the relative humidity was below 50 percent most of the time. The moisture content of rough rice can be re- duced to 4.7 percent (wet basis) when stored 42 days over a saturated solution of lithium chloride (74). The moisture content of rice can be reduced to about 6 percent with a silica gel drying unit.

PN-2771 Eice reduced to this moisture content probably FIGURE 8.—Plastic pipes 4 inches in diameter properly will retain viability for 5 years or longer when placed to provide semiautomatic watering system with water flowing from lateral at the right, into stored at room temperatures. individual nursery areas at the left. Between water- At Beaumont, Tex., seed of rice varieties, breed- ings, pipes are "capped" with plastic bags held in ing lines, and genetic stocks is dried in a silica gel place with rubber bands or other ties. drier to about 6-percent moisture and stored with silica gel in sealed 1.2-cubic-foot metal containers Bundles from large plots are threshed in a at about 34° F. About 2 pounds of silica gel is put Vogel thresher (fig. 16). Breakage caused by im- in each container. An indicator, such as cobalt proper adjustment of the thresher should be chloride crystals or paper, that changes color with avoided because it causes a bias of grain yield and increase in moisture is used, so that an increase in milling quality. moisture is easily detected. The silica gel is usually In some instances, bundles are placed on the changed every 2 years. Seed can be stored in poly- cross levees for a few hours after cutting and are ethylene bags with silica gel. However, because threshed the same day. The threshed grain is placed in cloth bags and artificially dried with forced air heated to between 90° and 100° F. Threshed samples usually are cleaned before weighing for yield determinations. It may also be necessary to break off awns and attached frag- ments of rachises. To break these off, place the sample in a section of automobile innertube and pound it several times. An easily cleaned aspi- rator, dockage machine, or small fanning machine can be used to remove foreign material. If- the threshed samples are relatively free from foreign material, the grain weight can be obtained before cleaning. Maintenance of viable seed of varieties and breeding lines is important in a rice-breeding pro- gram. The main causes of rapid deterioration of seed viability are moisture content of the seed and storage temperature. A method of drying the.seed and storage facilities with low humidity and low temperatures are needed. BN-22034 FIGURE 9.—Belt seeder, mounted on small tractor, used for The moisture content of the grain is probably seeding nursery yield experiments. RICE IN THE UNITED STATES 27

from commercial and introduced varieties, and (3) creation of new varieties by hybridization fol- lowed by selection. Irradiation breeding also has been used in the United States. The varieties now grown in the United States were developed by using progressively the three major methods. For example, the Gulfrose variety was selected from the progeny of a cross between a selection from an introduced variety and a pure line selection from a commercial variety.

Introduction The introduction of varieties from other coun- tries has been and still is an important source of germ plasm for lice breeders in the United States. Some of the introduced varieties have been de- veloped by breeders in the country of origin, and others have been indigenous varieties grown for many yeai-s by farmers in the country of origin. According to Jones (69), few varieties were in- troduced from about 1685 to 1889. It became evi- dent then that the varieties being grown in the United States were not as productive as varieties grown in other countries. Starting about 1890 and

BN-21977 FiGUEE 10.—Manually operated seeder used to seed selections. these bags do not exclude all the moisture, the silica gel must be changed more frequently. At Stuttgart, Ark., rice seed in cloth bags is dried to about 10 or 11 percent moisture and stored at about 70° F. and 70-percent relative humidity for 3 to 4 years and in bags or in envelopes in metal boxes. Rice seed is stored at about 40° and 60-percent relative humidity for long-term storage. At Crowley, La., and Beltsville, Md., seed is stored at a temperature of about 0° F. Seed stored in this manner has retained its viability for more than 20 years.

Breeding Methods The three major breeding methods commonly PN-2772 used for small grain have been used in rice breed- FIGURE 11.—Plastic jug with aerated cap and plastic flow- ing (69). These are (1) introduction of varieties control valve and tubing showing liquid nitrogen fer- from foreign countries, (2) selection of pure lines tilizer dripping into entering floodwater. 28 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

least 3 years in replicated yield trials and are eval- uated for disease reaction and cooking quality. When a variety is proved to be superior to com- mercial varieties, the seed is purified and increased and seed of the new variety is distributed to farmers. Selection Selections from introduced or commercial vari- eties have been an important source of rice varie- ties in this country. Practically all of the varieties grown in the United States from 1920 to about , .^.s^:,.^^':::y^^-'^f,m>'-{' 1945 were developed by this method. inm^^H. Jones (69) named the breeders who participated,

-.:.**K. in this work from about 1900 to about 1930. Most PN-2773 of these breeders were employees of the U.S. De- FIGURE 12.—Using drip method of applying liquid nitrogen partment of Agriculture. Chambliss and Jenkins fertilizer, one man can topdress 30 or 40 Vio-acre nurs- ery areas in 8 hours. (Sß) listed and described six varieties that were developed by selection. These are Acadia, Delitus, Evangeline, Salvo, Tokalon, and Vintula. These continuing to the present, many varieties have six varieties were not grown extensively but prob- been introduced. Jones (69) listed and described ably were important for a few years in local areas nine varieties that were introduced from other countries and grown in the United States. These were Carolina Gold, Carolina White, Early Wataribune, Honduras, Kyushu, Omachi, Onsen, Shinriki, and Wataribune. About 1935, Asahi was introduced from Japan by T. M. Sabora, a Texas rice farmer. He increased this variety for commer- cial production about 1940 (71). Thus, there have been at least 10 introduced varieties that have been grown rather extensively. None of these varieties are now of commercial importance. Varieties now introduced into the United States are grown the first year in a greenhouse away from rice production centers to avoid the introduction of diseases and insects. The seed produced in the greenhouse then is sown in a single-row plot at one of the rice experiment stations. Notes are taken on such characters as seedling vigor, straw strength, plant height, length of growing season, grain type, kernel characters, and reaction to dis- eases and insects. Whether the variety is a pure line or consists of mechanical or hybrid mixtures is also noted. Varieties that appear to have no desirable characters are discarded. Varieties that have some but not all desirable characters are held in reserve for possible future use in the breeding program. Varieties that look promising in all re- BN-22009 FIGURE 13.—Distributor used to apply nitrogen fertilizer spects in the preliminary trials are tested for at between the rows as topdressing. RICE IN THE UNITED STATES 29

increased, and it is released for commercial pro- duction. Most of the varieties developed by the selection method have been replaced by newer varieties, al- though there are a few notable exceptions. In 1963, Bluebonnet 50 was the leading variety in the South, and Eexoro was grown rather Avidely in Louisiana and Texas. Caloro was the leading vari- ety in California, and Colusa also was widely grown in that State. Hybridization Varieties developed by the selection method were a vast improvement over the varieties they replaced, but varieties developed by this method did not have all the desired characters. Since about 1922, rice breeders in the United States have used the hybridization method of rice breeding, and hundreds of crosses have been made. Three systems are followed in the hybridization method of breeding rice. These are: (1) Two vari- eties are crossed and the progeny grown in pedi- greed rows or bulk plots until pure lines are se- BN-22029 lected; (2) the backcrossing system in which two FiGUBB 14.—Force-feed distributor used to apply nitrogen varieties are crossed and the Fi plants crossed to fertilizer as a topdressing. one of these varieties and repeated for four or five times; and (3) the multiple crossing system in in Louisiana. Ten other varieties developed by which four or eight varieties are crossed in pairs, selection and listed and briefly described by Jones followed by crossing the Fi plants from the differ- (69) are Blue Rose, Caloro, Colusa, Early Prolific, ent combinations so that after two or three rounds Edith, Fortuna, Lady Wright, Nira, Rexoro, and of crossing, Fi plants are obtained that contain Shoemed. Other varieties developed by pure line selection are Conway (36), Sunbonnet (48), Blue- bonnet 50 (57, 71) and lola, Latex, Mortgage Lifter, Storm Proof, and Zenith [72), and Nova 66((5i). Selection is a relatively easy method for breed- ing rice, but it takes many years to develop a variety by this method. The procedure is to select a large number of plants or panciles from a variety that consists of diverse types. These selections are sown in single-row plots and carefully observed for the principal characters. If a selection from the progeny of a hybrid is segregating, desirable plants or panicles are selected and sown in single- row plots the second year. "When promising lines are breeding true to type, they are tested in repli- cated yield trials in the same manner as introduc- Bîr-2200! tions. When a selection is proved to be outstand- FIGURE 15.—Drying bundles of rice in trays with forced ing, it is named, the seed supply is purified and heated air. 488-871 O—73 3 30 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

toward the tip of the ñoret is opened, and an anther or two are placed in the emasculated flower. Usu- ally the anthers are broken open and the pollen is dusted on the stigma. This process is repeated until all flowers have been pollinated. After pol- lination, the panicle is again enclosed in the bag and the designation of the parents and the date of pollination are written on the tag. A modification of the clipping method has been used at Beaumont, Tex. In this method, the tips of the lemma and palea are clipped off at right angles to the longitudinal axis of the spikel&t. The anthers then are removed by vaccum, by means of a fine glass nozzle connected to a vac- uum pump that is driven by an electric motor operating from a storage battery. Emasculating is done in the early morning or in the late after- noon; and pollinating, bagging, and tagging are BK-21978 completed in the same manner as in the clipping FIGURE 16.—Threshing breeder seed with Vogel thresher. method described by Jones {69). Another technique that has been used for cross- some genes from each of the parent varieties. ing rice is a modification of a method developed These systems may also be modified in various for wheat and barley {105). In this method, the ways. lemma and palea are clipped at right angleë to the CROSSING TECHNIQUES.—The two basic tech- longitudinal axis of the spikelet at a point just niques used for making rice crosses are the clip- above the tip of the stigma. This cuts the an- ping and the hot water method. There are several thers, and it is not necessary to remove them. ijiodifications of the clipping method but in each The florets are clipped early in the morning or in case a part of the lemma and palea is removed. In the afternoon, and then the emasculated panicle the hot water method the panicle is emersed in hot is covered with a fairly large glassine bag. Later water, that day or the next day, pollen from a male An example of the clipping technique that has panicle that has many florets just starting to been widely used was described by Jones {69) and bloom is dusted on the stigmas of the emasculated may be summarized as follows: In the morning panicle. Before pollination the bottom of the before the rice begins to bloom, or in the after- inverted bag on the emasculated panicle is slit so noon after the daily blooming period has passed, that it can be spread open and can serve more or all except 10 to 20 spikelets are removed from the less as a funnel while the pollen is being applied. female panicle. The upper part of the lemma and The bag then can be closed, folded over, and palea of the remaining spikelets are clipped off at fastened with a paper clip. The pollinated panicle about a 45° angle. This removes about half of the is then tagged to show the designation of the par- lemma but only the tip or none of the palea so the ents and the date of pollination. anthers can be easily removed with fine-pointed As a result of the elevated temperature within forceps. The emasculated panicle then is tagged the bag, seed set may be poor in crosses made by and covered with a glassine bag. The same day or clipping the floret. In an effort to devise a tech- the next day between 9: 00 a.m. and 2: 00 p.m., de- nique for making a cross that would not require pending somewhat upon the weather, the emas- bagging the female panicles, Jodon (^5) devel- culated panicles are pollinated. Plants of the vari- oped the hot-water method for emasculating rice ety to be used as the male parent are examined, flowers. In this method, the female panicle is im- and a panicle that has spikelets about to bloom is mersed in water at a temperature of 40° to 44° C. taken. A spikelet that has the anthers pushing up for 10 minutes in the morning before any of the RICE IN THE UNITED STATEiS 31 florets have opened. This critical temperature reaction and seedling hardiness. In important renders the pollen grains ineffective but does not crosses where known genetic variables can be iden- prevent the normal functioning of the female por- tified, early-generation testing has proved very tion of the rice florets. Within a few minutes after effective. the panicle is removed from the warm water, the The rice-breeding procedures used in the F3 and florets that would have bloomed later that day subsequent generations usually are governed by open. These florets can be pollinated in the same particular circumstances such as available person- manner as in the clipping method. Spikelets that nel, physical facilities, financial considerations, already have been pollinated and spikelets that do urgency of a particular development, number of not open are clipped off. Spikelets at the base of characters involved, and mode of inheritance. the panicle usually are immature and can be left The pedigree system has been more widely used to produce mother seed. It may not be necessary than other systems of handling hybrid progenies, to bag the panicle, because the lemma and palea although modifications of the pedigree and bulk usually close in about 45 minutes. Thus, they systems are frequently followed. usually are closed before other plants start to shed In the pedigree system, the F2 plants are care- pollen. The pollinated panicle then is tagged to fully examined in the field and laboratory. The show the designation of the parents and date of selections that appear to be satisfactory for all pollination. characters studied are sown in individual rows SINGLE CROSSES.—The crossed seeds are planted (8 to 16 feet long and 12 inches apart) the in a greenhouse in the fall after the cross is made next spring. Seeds are spaced thinly in rows. or in the field next spring. When tihe Fi plants are Space seeding is seldom used for F3 and later grown in the greenhouse during the winter, a year generations. is saved because the F2 population can be grown in When facilities and circumstances permits, a the field the year after the cross is made. system of multiple screening of selections may be Usually several Fi plants are grown to assure used. Desirable plants are saved from the F2 that typical parental strains were used in making progenies in the field, and each plant is threshed the crosses. The cross should be made so that the individually. A small portion of the grain of Fi plant can be distinguished from the female each is milled, and plants having chalky kernels parent variety. For example, when pubescent X are discarded. Milled samples from the remaining nonpubescent varieties are crossed, the nonpubes- plants are tested in the laboratory to determine cent variety should be used as the female parent. cooking quality. Undesirable plants are eliminated, Fi plants should be propagated vegetatively until and the seeds of each of those remaining after the it is known that they are not needed for addi- double screening are divided into two parts. One tional seed production or other purposes. Vegeta- part (aibout 25 seeds) of each is seeded late in tive propagation of Fi plants is also helpful in the field under conditions believed to be favorable obtaining increased quantities of F2 seeds. for the development of blast. The other part is Seeds from the Fi plants are spaced thinly or seeded in the breeding nursery to observe agro- space sown in rows 12 inches or farther apart. nomic characters and to provide seed for further When space sown, seeds are placed 3 to 6 inches testing of lines surviving the multiple screening. apart in the row, so the F2 plants can be examined This screening method eliminates many undesir- individually for plant characters and disease able lines in the F3 generation. reaction. A further advantage of space seeding The Fa lines are carefully examined through- F2 populations is that all the seed produced by a out the growing season and usually from 6 to 15 selected F2 plant can be harvested and used for panicles are selected from lines that have the (1) preliminary quality tests on milled samples desired plant and grain type and that appear to (5 to 10 grams), (2) growing at different loca- be resistant to diseases. The quantity of seed tions or dates of seeding to obtain information from each panicle is seldom sufficient for seeding on reaction to diseases and to photoperiod and at more than one or two locations. Small samples other environmental influences, and (3) seedling for quality tests and a few seeds for disease nur- tests in greenhouse or growth chamber for blast series may be obtained by bulking part of the seed 32 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

from each of the 6 to 15 panicles harvested from the next year. This procedure is followed for a given Fg line. The remaining panicles in a line each succeeding generation until Fe or F7. In may be harvested in bulk to provide additional bulk-hybrid populations, special care should be seed for testing the line for resistance to straight- taken to be sure that desirable plant types are not head and blast and for seedling hardiness, grain being eliminated through natural selection. For quality, or other characters. The F4 lines are han- example, tall and profuse-tillering types tend to dled in much the same manner as the F3 lines. eliminate short types (8). Usually from 3 to 6 selected panicles of each line The bulk system often is modified in various are sown in panicle-row plots the following year. ways. For example, if there is a differential plant Early-generation testing, is continued until it is reaction to a natural or artificially induced injec- reasonably certain that the lines are no longer tion of a serious disease, a bulk is made up by se- segregating for the characters under study. From lecting the resistant plants. Plants that make up 10 to 15 panicles are selected from nonsegregat- the bulk are sometimes tested for processing and ing rows, and then the rows are harvested and cooking quality. Where the parents differ in grain threshed in bulk. Part of the seed from the type, panicles having the desired grain types are harvested rows is used for more extensive labora- selected in bulk or the bulked seed is graded on the tory grain quality tests. basis of length and width, and the desired grain Lines that have suitable grain type are then type is saved. This operation might be repeated tested in a preliminary yield trial. Some lines for several generations. When unfavorable weather may be discarded because of lack of vigor, weak or a shortage of help makes selections in the field straw, disease susceptibility, undesirable cooking impossible, a composite may be harvested and quality, or other undesirable characters. Lines that panicles selected in the laboratory. In this case, the are heterozygous for one or more characters but grain from selected plants is combined for seeding otherwise appear to be desirable are replaced by the bulk plot the next year. pedigree sublines selected from later generation In California, progenies from each cross are panicle rows. True breeding lines usually can be sometimes grown in small, water-seeded plots isolated in six to eight generations. The true breed- where the material is seeded under conditions sim- ing lines then are tested for at least 3 years to ilar to those of the commercial ricefields. This determine yielding ability, disease resistance, time method aids the breeder in selecting plants that of maturity, plant height, stiffness of straw, and emerge readily through deep water. Also, when cold tolerance, and to check thoroughly milling, rice is water seeded, some varieties do not develop processing, cooking, and other characters. Lines sufficient roots, and plants may tend to lean as the that prove to be superior to existing varieties in grain matures. Thus, when rice is water seeded in one or more characters are then named, the seed breeding studies, plants may be evaluated for re- is purified and increased, and the seed of the new sistance to this type of lodging. variety is distributed to growers. Each year the material may be subjected to The bulk system has been used to a limited natural selection pressures in an attempt to elimi- extent, particularly where time and space were nate undesirable types. Examples of natural selec- limiting factors. True breeding lines usually can tion pressure are: Harvesting before late plants be obtained in fewer generations when the pedi- are mature if short-season types are desired ; seed- gree method is used. ing under conditions where straighthead is likely In the bulk system of breeding, the Fi plants to occur; or inoculating with Aphelenchoides tes- are grown and the F2 populations are sown in the seyi Christie (sometimes called A, oryzae Yokoo) same manner as in the pedigree method. The F2 to eliminate plants susceptible to white tip. After population is harvested and threshed in bulk. six or seven generations, a large number of plants Progenies from each cross may be kept separate or panicles are selected, preferably from a space- or the progenies from several crosses with similar planted population. The plants are carefully ex- parentage may be combined. From 10 to 25 rod amined and only those that appear to have strong rows are grown in F3. They are harvested and straw, disease resistance, and good grain type are threshed in bulk and down on a similar size plot selected. Seeds of the selections are then examined RICE IN THE UNITED STATTE^S 33 in the laboratory much the same as seeds from F^ Irradiation plants are examined, and those that appear to have Irradiation breeding has been used in the United the desired kernel characters are sown in single- States. Several mutants that may be useful in row plots the next year. The selections are then breeding programs have been obtained in this handled in the same manner as advanced-genera- manner, but no varieties have been developed. tion selections obtained by the pedigree method. Maliy rice varieties have been developed in the Breeding for Agronomic Characters United States by the hybridization method. Vari- Eice often is seeded early in the spring when the eties that were named and released are as follows : soil temperature is low, or it may be sown in water Arkrose, Belle Patna, Bluebelle, Bluebonnet, Cal- that is below the optimum temperature. Thus, seed- ady, Century Patna 231, Cody, CS-M3, Dawn, ling vigor is an important character of rice varie- Delia, Delrex, Gulfrose, Improved Bluebonnet, ties, and it is desirable to study seedling vigor and Kamrose, Lacrosse, Magnolia, Missouri E-500, cold tolerance concurrently and to combine the two Nato, Northrose, Nova, Prelude, R-D, Rexark, characters in a rice variety. R-N, Saturn, Starbonnet, Texas Patna, Toro, TP At optimum temperatures, indica varieties gen- 49, Vegold, and Vista. Many of these varieties were erally grow faster in the seedling stage than do not in commercial production in 1970, although japónica varieties. However, some of the japónica varieties developed from crosses comprised about varieties are more tolerant to low temperatures, 90 percent of the U.S. rice acreage that year. and seedlings of these cold-tolerant japónica varie^ BACKCROSS.—The backcross system or a modifi- ties may grow and develop faster than seedlings of cation of the typical backcross system is success- indica varieties when the temperature of the soil fully used to some extent in rice breeding in the or water is low. United States. As with other small grains, this Vigor and cold tolerance of seedlings are system is used when it is desired to transfer one studied under controlled conditions {6^88) and in gene or a small number of genes to a fairly well- the field. In the controlled studies, the seeds are adapted variety. This system has not been used as sown in water and the temperature is maintained widely with rice as with other small grains. Com- at 60° F. The length of longest leaf at about 30 mercial rice production methods in the United days is the criterion used to judge tolerance and States have undergone periodic changes (combine vigor. The varieties and lines that are vigorous harvesting, increased fertilizer use, new herbi- and cold tolerant under controlled conditions are cides) that have resulted in rather drastic changes tested in the field. Field experiments are con- in plant-type requirements. Because rice breeders ducted by water seeding in fields where the irriga- have had to search for divergent types, the use of tion water is cold, or by drill seeding early in the standard varieties as recurrent parents in a formal spring. Because variability is high under field backcross program has been prevented. As produc- conditions, results for a particular experiment are tion methods become more stable and satisfactory not always dependable. But over a period of plant types are developed, and as rice breeding years, useful information can be obtained on the progresses in the United States, the backcross performance of varieties under these adverse method may be used more generally. Adair, Miller, conditions. and Beachell (9) described briefly the develop- Varieties also differ in their response to low ment of hoja blanca resistance, long-grain types by temperatures during the ñowering stage. This backcrossing. character of rice varieties and breeding lines is Calrose (71), a medium-grain variety, was de- studied by seeding late, so that pollen develop- veloped by this method of breeding. ment and flowering occur when the temperatures MULTIPLE CROSS.—Multiple crossing has been are low. In Texas, when temperatures are below used very little by rice breeders in the United the optimum in October when rice is flowering, States. Several varieties have been developed from hardy varieties, such as Caloro, continue to de- rather complex crosses, but none of these were se- velop; whereas cold-sensitive varieties, such as lected from a population obtained from systemati- Bluebonnet 50 and Century Patna 231, may not cally crossing four or more varieties. produce seed. 34 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE.

Rice varieties also vary in response to high Nato, Nova 66, and Dawn. The long-grain varieties concentration of salts in the irrigation water and Bluebonnet 50, Starbonnet, and Belle Patna Imve in the soil. Varietal response to salt was studied shown somewhat less tolerance, and the very short at Stuttgart, Ark., and Beaumont, Tex. At season variety Bluebelle often has performed very Stuttgart, on soils with a high salt content, Ark- poorly when seeded on soils with a pH above 7.0. rose produced 4,180 pounds per acre and Blue- Dawn has consistently shown more tolerance to bonnet 50 produced 3,172 pounds. On soils with alkaline soils than other commercial long-grain lower concentration of salts, Arkrose produced varieties. 4,531 pounds per acre and Bluebonnet 50 pro- Varieties that produce maximum yields must duced 3,838 pounds. At Beaumont, with a con- have relatively short and sturdy straw, so that the centration of sodium chloride of about 2,500 parts rice does not lodge before harvest (fig. 17). per million in the irrigation water, Caloro pro- Japónica varieties have smaller and shorter culms duced 1,789 pounds of grain per acre, whereas and shorter, narrower, and darker green leaves Toro produced only 842 pounds. When no salt was added to the irrigation water. Caloro produced than most indica varieties have. Although there 4,820 pounds per acre and Toro 3,716 pounds. are exceptions, japónica varieties generally pro- The Beaumont studies were conducted in one- duce higher yields than indica varieties produce; tenth-acre blocks of Beaumont clay soil. Approxi- this indicates that morphologic characters might mately 2,500 parts per million of salt was main- be associated with grain yield. The japónica vari- tained in the irrigation water by adding sodium eties have tough but willow Stems. The less willowy chloride. Frequent additions of salt were needed straw character of the indica varieties is preferred because of dilution from rains and soil moisture because this type is less likely to lodge. exchange. As much as 15,000 pounds of salt per For several years, rice breeders have been acre was used during the 70- to 80-day test. Salt searching for plants with short, slender stems; was first added to the irrigation water when the with erect, relatively narrow, dark-green leaves rice plants were about 45 days old. of intermediate length ; and with low percentages Under these conditions an attempt was made to of sterility at high rates of nitrogen fertilization. grow breeding lines in single-panicle rows as well Strains approaching this description have been as to conduct yield tests in replicated nursery selected from crosses between a Taiwan variety, plots. Widely different responses occurred because Tainan-iku No. 487 (P.I. 215,936),= and United of weather conditions during blooming and im- mediately thereafter. When blooming occurred ^ P.I. refers to the plant Introduction accession number during periods of cloudy weather and moderate of tlie Agricultural Research Service, U.S. Department of temperatures, practically all lines showed good Agriculture. seed sets. When temperatures were high and humidity was low, practically all lines showed poor seed sets. After several years, the test was abandoned because it required a great amount of time and effort and provided little information. In more recent experiments at Stuttgart, nu- ^.^afr^rfU.« <. '^ merous rice varieties have been grown in yield trials or reaction tests, or both, to determine their response to high pH (7.2 to 8.0) soils. Varieties grown on high pH soil produced only 61 percent as much grain in 1956 or 65 percent as much grain in 1966 as when the same varieties were grown on lower pH (5.6 to 7.0) soils. Ctirrently grown rice varieties that have shown appreciable tolerance BN-22021 to alkaline soils to date include Arkrose, Caloro, FiGTJEE 17.—Field of Caloro rice showing serious lodging. RICE IN THE UNITED STATEIS 35

State varieties. Ten short-stature selections from panicles of this selection were normal in length these crosses were grown in a variety-fertilizer and floret number, and the uppermost internode test in 1961. They averaged 4^224 pounds of was of about normal length. The lower internode rough rice per acre at the 160-pound nitrogen of the Century Patna 231 selection was of about rate and 3,367 pounds at the 80-pound rate. The normal length. The lower internodes were only highest yielding strain yielded 5,658 pounds at slightly elongated so this selection had extremely the 160-pound rate and 4,022 pounds at the 80- short and sturdy culms. pound rate, compared with Bluebonnet 50, an The grains of the dwarf were of normal length indica type variety, which yielded 3,448 and 2,916 but appeared to be slightly tapered. This dwarf pounds, respectively, at the two rates of nitrogen was crossed with normal Bluebonnet 50, Century fertilization. Patna 231, Eexoro, and Texas Patna. Dwarf type In an effort to find short-stature types possess- selections resembling the original strain were re- ing superior yielding capacity, varieties that com- covered in about a 3 normal : 1 dwarf-type ratio bine extremely narrow and erect leaves and small in all crosses. The original and the dwarf-type stems have been crossed. selections from the crosses were tested in yield Short-strawed plants have been found in hybrid experiments and invariably showed about 15- to and irradiated populations, as mutations from 20-percent lower yields than normal strains varieties, and in introductions from foreign coun- showed. All the dwarf-type strains showed consid- tries. In Arkansas, naturally occurring dwarf erably more sterility than the normal strains types were saved from C.I. 9187,^ a high-yielding, showed but this was probably responsible for the lodging-resistant, early long-grain experimental reduced yields. The cause of the increased sterility variety that has short straw and narrow leaves. was not determined. This promising dwarf type Crosses were made with Bluebonnet types to im- will be studied further. prove milling quality, and dwarf and semidwarf Generally a high rate of nitrogen fertilizer is types were saved for testing. applied to rice-breeding nurseries to select types The short-stature types showed considerable that respond to fertilizer and that resist lodging. variation in plant height because of the degree of When pure lines are isolated, they are then tested internode elongation. Club and grassy dwarfs at two rates of nitrogen fertilization. Usually a seldom grow more than 12 to 18 inches tall. Since randomized, split-plot design is used with varie- grains are shortened and frequently are otherwise ties as the main plots and with nitrogen rates a3 distorted, they are of no economic value. Some subplots (ñg. 18). intermediate-height, dwarf-type lines exhibit At Stuttgart, Ark., a continuing experiment de- varying degrees of grain distortion. Other inter- signed cooperatively by the agronomist, the plant mediate-height, dwarf-type lines appear to pro- breeder, and the plant pathologist has been estab- duce grains of normal size and shape. lished to compare outstanding experimental va- All of the dwarf-type plants examined ap- rieties with commercial varieties they might be peared to have the normal number of nodes. The expected to replace. Varieties are tested in matu- rity groups and receive nitrogen fertilizer at a reduced height was governed by the extent of in- number of rates and times. The varieties are ternode elongation, including the uppermost in- checked closely for grain yields and disease reac- ternode. In all but one dwarf-type selections exam- tion, lodging, and other characters in the field. ined, all internodes of the stem, including the pe- Milling yields and chalkiness of the grain and duncle, showed reduced elongation when compared kernels resulting from the various treatments are with normal height plants. An exception to this also checked. To develop a satisfactory nitrogen elongation pattern was observed in a dwarf plant fertilization program for each variety so that it selected at Beaumont, Tex., in 1955, from an irra- will give consistently high grain yields with a diated population of Century Patna 231. The minimum of lodging and disease, possible new va- rieties are tested 2 or 3 years as part of a final ® C.I. refers to the cereal accession number used by the Agricultural Research Service, U.S. Department of evaluation before their release. Agriculture. In fertilizer-variety tests at Stuttgart, Ark., C.I. 36 AGRICULTURE HANDBOOK NO. 289, U.S. DEFT. OF AGRICULTURE. 9 percent greater for Starbonnet than for Blue- bonnet. Starbonnet and Bluebelle, which have relatively short, stiff straw and fairly erect plant type, can effectively use midseason nitrogen at an earlier stage of plant development than somewhat taller, leafier varieties such as Nova 66 can use it. Notes on breeding lines and strains in yield tests recorded about 30 and 60 days after seeding and at maturity provide valuable information on plant type. By using symbols, brief notes are obtained on (1) habit of growth of leaves (erect to spreading) ; (2) color of leaves (light to dark green) ; (3) leaf width (narrow to wide) ; (4) plant or seedling height (short to tall) ; and (5) degree of tiller- ing (low to high). Calculating sterility percentage by counting the number of sterile and fertile florets on selected panicles of breeding lines appears to have merit. The grain yield of breeding lines and experi- mental varieties is determined by using replicated BN-22008 nursery plots. For comparison, standard varieties FIGURE 18.—Bluebonnet 50 fertilized at two rates of ni- trogen fertilizer: Left, plot fertilized with a total of are included in each trial. The experimental design 160 pounds per acre of nitrogen (divided into applica- is a complete randomized block, usually with four tions of 80 pounds at seeding, 40 pounds at 62 days after replications. The plots are about 1 rod (5 meters) seeding, and 40 pounds at 100 days) was 53 inches tall long and three or four rows wide. Usually the and produced 4,892 pounds of rice per acre ; right, plot rows are spaced 9 inches apart in the four-row receiving no nitrogen was 40 inches tall and produced plot or 12 inches apart in the three-row plot. The only 2,214 pounds of rice per acre. plots are trimmed to a uniform length before harvesting. For yield determination, the center 9434, a very short-season, experimental long-grain row is harvested in the three-row plot, and the variety, produced 5,992 pounds of grain per acre two interior rows are harvested in the four-row plot. One or both of the border rows may be har- compared with 5,839 pounds produced by Blue- vested in order to have a larger sample for milling bonnet 50. In both tests the treatment included 200 and cooking tests. pounds of nitrogen per acre, but C.I. 9434 pro- An alternate method is the use of four-row duced this relatively high grain yield in 45 days plots usually 15 or 16 feet long with a 12-inch less time than Bluebonnet 50; and the straw of spacing between rows. An 8-foot segment from the C.I. 9434 was 14 inches shorter. two center rows is harvested. By careful harvest- The results from these experiments indicate ing, the remainder of the plot is left standing for that it will be possible to develop short-strawed, later observations and for collection of panicle nonlodging varieties that respond to high rates samples. of fertilizer. Progress made to date includes the In California, Avhere the rice on all farm fields development of the shorter strawed varieties Blue- is sown broadcast in the water, the method used belle and Starbonnet {28, 64). Starbonnet plants for testing varieties is somewhat different. The produce more tillers than Bluebonnet 50, usually rice is sown in the water by hand. have narrower and shorter leaves, produce pani- The design used for preliminary trials is a com- cles that are shorter and less dropping, average 15 plete randomized block with three or four repli- percent shorter in height, and are much more re- cations. The individual plots are a single row, sistant to lodging. Because of higher field and 12 to 16 feet long, with the rows spaced about 24 mill yields, the per acre value of milled rice was inches apart. The preliminary trial gives infor- RICE IN THE UNITED STATEIS 37 mation on the ability of the variety to emerge and palea at the widest point (fig. 19, B). Width through the water, straw strength, and other char- for brown and milled rice is the distance across acters. The best varieties then are tested in ran- the kernel at the widest point (fig. 19, A). domized, quadruplicated plots large enough to be harvested with a small combine. Time of maturity (length of growing season) is an important consideration in the breeding pro- gram. Although early and very early types have gained favor in recent years, a complete range of maturity types is maintained. Partial dormancy in rice is desirable, so that rice does not germinate if rain and humid weather occurs at harvesttime or if plants are badly lodged. Seed of nondormant varieties will germinate be- fore cutting if there is a prolonged period of rain after the grain is ripe. It sometimes is desirable to break dormancy of seeds soon after harvest. This is necessary in the case of crossed seeds to be planted in the greenhouse in the winter or other breeding material to be grown during the off sea- son in another region. This can be done by treat- ing in a 0.10- to 0.05-percent solution of sodium hypochlorite for 24 hours or by heating in shal- low, open containers for 3 to 5 days at 47° to 50° C. Rice varieties in the United States are divided by grain size and shape into three types. These are short- (Pearl), medium-, and long-grain types. Examples of each type are Caloro, Nato, and Blue- bonnet 50, respectively. The grain type can be vis- ually classified ; but for more critical comparisons of varieties and for classification, more exact meas- urements are needed. At the Regional Rice Qual- ity Laboratory at Beaumont, Tex., the various grain types are characterized objectively accord- ing to length, width, length/width ratio, thickness, and grain weight. Dimensions of rough (paddy), brown, and milled (head) rice grains are meas- ured and reported in accordance with the follow- ing definitions : (1) Length of awnless rough rice is the straight- line distance (millimeter) from the point of disarticulation of the grain, which is below the outer glumes, to the tip of the apiculus (fig. 19, B). For awned rough rice the tip of the lemma is the reference point. Length for brown and milled rice is the distance between the most distant tips of the kernel, including the embryo of the brown

rice kernel (fig. 19, A). BN-21515 (2) Width (dorsiventral diameter) for rough FIGURE 19.—Points from which rice grain and kernel rice is the distance (millimeter) across the lemma measurement should be made. 38 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE,

(3) Thickness (lateral diameter) for rough rice millimeter lens. A millimeter scale was placed is the distance (millimeter) from one outside sur- on the glass slide. The image of the millimeter face of the lemma to its opposite at the thickest scale was magnified 10 times and made possible point (fig. 19, D). Thickness for brown and milled accurate focusing before use. The adjustable rice is the distance from one side of the kernel to bellows to which the 50-millimeter lens was at- its opposite side at the thickest point (ñg. 19, C). tached made it possible to obtain a clear-cut By modifying a photographic enlarger, a device image. The enlarged images of the rice grain was built for measuring the length and width of were measured to obtain the length and width. rice (fig. 20). The enlarger was mounted on a box The thickness of the rice grain was determined with three sides that were painted black on the by using a micrometer caliper graduated in milli- inside. A black cloth was placed over the open meters. The micrometer was placed in a small side of the box to exclude most of the light, so vice attached to a table. A sheet-metal tray at that a distinct image of the rice grain would ap- the base of the micrometer, to catch grains pear on the white grid. Minute adjustments were dropped during the operation, is useful. made to obtain a magnification of exactly 10 times. If similar equipment is not available, a random The light source was a 6- or 12-volt cold, sealed- sample of 10 grains or kernels can be placed adja- beam, automobile spotlight. The spotlight had a cent on transparent tape in the desired position smooth surface, so that light or shaded areas would for the particular measurement, and the total not be projected on the grid. Aiiy cold light source length, width, or thickness can be measured with of similar intensity should be satisfactory. a fair degree of accuracy by using a transparent ruler. A 50-mm.F/4.5 liminized lens was used. This The length, width, and thickness of milled lens made it possible for the enlarged image kernels are determined by measuring 20 whole (lOX) to be projected on a grid that was less kernels selected at random from a representative than 24 inches from the lens. sample. The coefficient of variation for each di- A glass slide that held 10 grains was placed in mension is calculated for each of 20 kernels to the light field just above the bellows and the 50- determine the uniformity of size and shape. Size and shape classifications for brown kernels are as follows :

Average length Average length/ (Millimeters) width ratio Long (D- 6.61 to 7.5 over 3. Medium (M)... 5.51 to 6.6 2.1 to 3. Short (S) 5.5 or less Up to 2.1. The average length, length/width ratio, thick- ness, and 1,000-grain weight of rough, brown, and milled (head) rice for each of 18 varieties are tabulated in table 6.

Testing for Milling, Cooking, and Processing Qualities The determination of milling, cooking, and processing qualities of hybrid progenies, breeding lines, and new varieties is an essential part of the rice-breeding program. New varieties that are released for commercial production must meet

BN-22019 established standards for these qualities. Certain FiGTiBE 20.—Device for measuring rice grains and kernels. cooking and processing qualities are historically RICE IN THE UNITED STATES 39

TABLE 6.—Grain characters for 18 rice varieties

Grain characters ^ Grain type and variety Grain form ^ Length L/W Thickness 1,000-grain weight

Short-grain : Millimeters Millimeters Grams Rough- 7.4 2.1: 2.2 28 Caloro Brown _ 5.4 1.8: 2.1 24 .Milled. 5.2 1.7: 2.0 22 'Rough- 7.5 2. 1: 2.3 30 Colusa ^. Brown- 5.4 1.8: 2.1 24 Milled- 5.3 1.8: 2.0 23 Medium-grain: Rough- 8.4 2.6:1 2.2 30 Arkrose Brown. 6.3 2.3:1 2.0 24 .Milled. 5.8 2.2:1 1.9 22 'Rough- 8. 1 2.5:1 2.0 25 Calrose Brown. 6. 1 2.2:1 1.9 21 ^Milled- 5.7 2.1:1 1.8 20 'Rough_ 7.9 2.6:1 2. 1 25 CS-M3 Brown - 6.0 2.4:1 2.0 22 .Milled. 5.7 2.2:1 1.8 21 'Rough - 8.0 2.6:1 1.9 23 Nato Brown. 5.9 2.3:1 1.8 18 .Milled. 5.5 2.3:1 1.7 17 'Rough- 8.2 2.6:1 2.0 26 Nova 66 Brown. 6.1 2.2:1 1.9 21 .Milled. 5.7 2.2:1 1.8 19 Rough. 8.0 2.6:1 2.0 24 Saturn Brown- 5.9 2.3:1 1.8 19 .Milled. 5.7 2.3:1 1.7 18 Rough- 8.5 2.7:1 1.9 25 Zenith- Brown . 6.3 2.4:1 1.8 20 .Milled- 5.9 2.4:1 1.7 19 Long-grain : ''Rough_ 9.4 3.9:1 1.8 22 Belle Patna Brown _ 7.3 3.4 1.7 18 .Milled- 6.7 3.4 1.6 17 'Rough- 9.6 3.9 1.9 24 Bluebelle Bro wn - 7.5 3.5 1.8 19 .Milled- 6.9 3.4 1.7 18 ''Rough- 9.5 3.9 1.9 24 Bluebonnet 50 Brown- 7.4 3.6 1.8 20 ,Milled_ 6.9 3.5 1.7 18 ^Rough- 9.3 3.9 1.8 22 Dawn Brown- 7.4 3.6 1.7 18 ,Milled- 6.8 3.6 1.6 16 ^Rough- 9.6 3.9 1.9 23 Delia (aromatic)^ Brown- 7.5 3.6 1.7 19 .Milled. 7.0 3.6 1.6 18 'Rough. 9.2 3.9 1.8 21 Rexoro.. Brown . 7.0 3.5 1.7 17 .Milled- 6.7 3.5 1.6 15 'Rough- 9.2 3.9 1.8 22 Starbonnet Brown - 7.2 3.6 1.6 17 .Milled. 6.7 3.5 1.5 16 See footnotes at end of table. 40 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

TABLE 6.—Grain characters for 18 rice varieties—Conimnoia

Grain characters 2 Grain type and variety Grain form ^ Length Ratio L/W Thickness 1,000-grain weight

Millimeters Millimeters Grams (Rough- 9.2 3.8:1 L8 22 Brown. 7.1 3.4:1 1.7 18 Milled- 6.5 3.3:1 L6 16 fRough- 9.5 3.9:1 L9 22 TP49 ^Brown_ 7.3 3.6:1 L7 18 IMilled. 7.0 3.5:1 L6 17

1 Rough=unhulled grain; brown=grain with hull removed; milled=whole grain milled kernels with hull, bran, and germ removed. 2 Average values for fully developed mature grain.

associated with specific grain types. For example, process are performed mechanically with a mini- most of the long-grain varieties grown in the mum amount of manual labor, including the trans- United States cook dry and flaky, and some are fer of rice from one machine to another for the used for specific processed products. The short- next series of operations. Fraps (37), Geddes (38^ and medium-grain varieties grown in the United pp. 20.Í3-2051). Kik and Williams {76), Wayne States are more moist when cooked than are the (101), and, more recently, Kester {75) and Witte long-grain varieties and are used for specific proc- {109) have described the commercial milling of essed products such as dry cereals. Thus, it is rice as practiced in the United States. Additional desirable that a new variety have the same cook- information describing rice-milling and proc- ing and processing qualities as the variety it essing equipment has been published by the Food replaces. and Agriculture Organization of the United Na- tions {11,30). Milling Quality The milling quality of rice is based on the yield The objective of rice milling is the removal of of head rice obtained because head rice is usually hulls, bran, and germ with a minimum breakage the milled product of the greatest monetary value. of the endosperm. The milling process generally Yields of head rice vary widely, depending on consists of four fundamental operations: (1) variety, grain type, cultural methods and other Cleaning the field-run rough rice to remove such environmental factors, and drying, storing, and things as mud lumps, rice stems and leaves, milling conditions. The yield of total milled weed seeds and stems, and other foreign matter; kernels (head rice and all sizes of broken kernels) (2) shelling the cleaned rice to remove the hulls; is important, too, and this yield is influenced by the (3) scouring the brown rice to remove the course proportion of hulls and the amount of fine par- outer layers of bran, white inner bran, aleurone ticles of broken kernels unavoidably included in layers, and germ; and (4) grading the mixture the bran fraction during the milling process. of whole and broken milled kernels according to In rice-breeding programs, rigid laboratory size classes known as head rice (whole-grain milling tests are required to insure that any new milled kernels) second head (larger pieces of variety released will consistently produce high broken milled kernels), screenings (smaller pieces yields of head rice and total milled rice. At the of broken milled kernels), and brewers rice (very Cooperative Rice Quality Laboratory at Beau- small pieces of broken milled kernels) (9^ 18). In mont, Tex., three methods are available for esti- modern rice mills, all operations in the rice-milling mating the milling quality of rice varieties and RICE IN THE UNITED STATES 41 selections. The method used depends on the plate. Whole kernels drop off the end into a con- amount of rice available for testing. The official tainer ; broken kernels fall into the indents in the grading method for determining the milling qual- plates. Plates with specific size indents are useà ity of rough rice (by United States standards, for each grain type. Results usually are reported described by Smith {95-98) and adapted to labora- as percentage of hulls, bran, head rice, and total tory conditions, gives the breeder comparative in- milled kernels. formation regarding milling quality of the more A method also is available for milling very advanced selection^ grown in larger plots. The small samples of rice {92). This method can be method requires 1,000 grams of rough rice for used to mill the rice from one panicle. The rough each determination. rice sample is thoroughly cleaned and hulled in a A modification of the official method requires McGill sheller, and a 5-gram sample is weighed only a 125-gram sample {21) to estimate the for milling. The weighed sample is put into a milling quality of rice and enables the breeder to test tube with 3 grams of abrasive (40- to 60-mesh check milling quality at very early stages of selec- fused white aluminum oxide or clean, sharp, tion. The samples of rough rice are sealed in white quartz sand). The stoppered test tube is glass jars and kept at room temperature for at mounted in the test-tube miller, which holds 80 least 24 hours to bring the moisture content to test tubes, and is shaken for 45 minutes at a speed equilibrium before the rice is milled. The mois- of approximately 390 strokes per minute. The sam- ture content is determined, a 125-gram sample of ples then are removed and polished in a small the rice is hulled in a McGill sheller, and the sample polishing machine {93). This method of brown rice is weighed. The weight of the hulls milling small samples not only gives information is determined by subtracting the weight of the on the milling quality of individual plant selections brown rice from the weight of the rough rice. but also provides a milled sample to use for pre- The brown rice sample is milled in a No. 1 or No. liminary tests of the physical and chemical prop- 2 McGill miller for 30 seconds, using a 14-pound erties of the kernel. weight over a steel plate that covers the rice. The The average head and total milled rice yields sample is immediately milled a second time for for each of 18 varieties grown in the Uniform Rice 30 seconds, using a 4-pound weight. This second Performance Nurseries in Arkansas, Louisiana, milling gives a high polish to the rice kernels but Mississippi, and Texas from 1960 through 1969 causes little additional breakage. The milled ker- are tabulated in table 7. Estimated yields were nels are sealed in a glass jar to prevent unneces- determined according to the modified procedure sary breaking due to rapid or uneven cooling. The of Beachell and Halick {21). bran is screened through a 20 X 20 mesh wire screen to recover the small, broken kernels that Cooking and Processing Qualities passed through the miller screen. The finely broken kernels thus recovered are aspirated and Eice varieties differ greatly in cooking and added to the milled kernels. The milled kernels processing qualities. Among the domestic varie- are weighed when the rice has cooled to room ties, the quality of home-cooked rice has been temperature. The weight of the bran and polish described as varying from very sticky to flaky is the difference between the weight of brown rice {53). Fully cooked grains of typical United States short- and medium-grain varieties are and the weight of total inillèd rice. usually somewhat sticky, relatively firm, and tend The whole kernels are separated from the total to stick together. Typical long-grain varieties milled portion with a sizing device developed by usually cook to a flaky state with a minimum of the Gran Division, Consumer and Marketing splitting and do not tend to stick together. Other Service, U.S. Department of Agriculture. This terms used to subjectively described cooking qual- devices makes use of two indented plates, with flat- ity are moist or dry, soft or firm, and mealy or bottom holes, tilted at a slight angle and shaken chewy. Since different cultural groups prefer by an eccentric mechanism. During the shaking different textures, there is a rather widespread de- motion, the rice travels the length of the top mand for all types for use as home-cooked table sloping plate and drops onto the bottom sloping rice. 42 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE There is also a demand for all types of rice for and broken rice is often used in brewing. Typical use in the widely different prepared products. long-grain varieties are preferred for many par- Processors of rice prefer different textures for boiled and quick-cooking products, and specific their various products and also specific qualities long-grain varieties are preferred for certain adapted to the processes themselves. According canned soup products. Medium- and short-grain to Kester (75), a substantial amount of the do- varieties are more suitable for dry breakfast mestic rice crop is processed into various kinds of cereals and for use in baby foods and in brewing. prepared foods such as parboiled rice, quick-cook- The short-grain types exclusively are used for ing rice, breakfast cereals, canned rice, canned making puffed rice. soups, canned rice and vegetable mixtures, dry Although in the United States each grain type soup mixes, enriched baby foods, and frozen is generally associated with specific cooking and dishes. Eice flour is used in various processes. processing qualities, notable varietal exceptions

TABLE 7.—Milling yields for 18 rice varieties grown in Uniform Performance Nurseries in Arkansas, Louisiana, Mississippi, and Texas, 1960-69 [Values are averages and are for first cutting only]

Milling yields ^ Years Grain type and variety tested Head[ rice Total milled rice (average) Average Range

Short-grain: Number Percent Percent Percent Caloro 6 63 39-72 74 Colusa 8 66 29-73 74

Average 65 . 74

Medium-grain : Arkrose 6 60 31-70 71 Calrose 6 68 60-71 72 CS-M3 2 66 58-72 71 Nato 10 69 63-73 72 Nova 66 9 63 44-70 69 Saturn 8 65 45-72 72 Zenith 10 64 44-71 70

Average 65 . 71

Long-grain: Belle Patna 10 56 43-69 70 Bluebelle 9 60 51-70 70 Bluebonnet 50 __ 10 59 44-71 71 Dawn 8 54 38-68 68 Delia (aromatic). 9 57 39-68 70 Rexoro ^ 4 56 44-65 69 Starbonnet 6 61 45-69 68 Toro 10 64 54-70 70 TP49 2 4 57 52-62 70

Average 57 . 70

^ Head rice is the whole-grain milled kernels. Total milled rice is the head rice and all sizes of broken kernels. Mill yields are based on clean rough rice samples. 2 Grown only in Louisiana and Texas. RICE IN THE UNITED STATES 43 within each grain type have been reported; and nels in contact with dilute alkali were reported the nontypical cooking and processing quality of more recently {79), Two distinct reactions were these grain-type exceptions in relation to meas- noted: (1) Spreading where the kernels disinte- ured differences in some chemical and physical grated into small granules and spreading to sev- qualities of the rice grain has been discussed eral times the original kernel size; and (2) clear- ing where the starch is solubilized with a loss of In rice-breeding programs, cooking and proc- opacity. Spreading and clearing were evaluated essing quality is considered to be an important on a numerical scale from 1 (minimum) to Y measure of the suitability of a variety or selection (maximum). A slight-to-moderate reaction was for specific purposes. At the Cooperative Eice characteristic of most domestic long-grain varie- Quality Laboratory at Beaumont, Tex., results of ties and a more pronounced reaction was charac- specific chemical and physical tests collectively teristic of most short- and medium-grain varieties. serve as indices of cooking and processing qual- A very high correlation between the alkali reac- ities. The results guide the rice breeder in select- tion and gelatinization temperatures has been ing lines that combine the desired cooking and observed {^0). processing qualities and agronomic features. The The pasting quality of several domestic varieties merit of this type of evaluation has been clearly as determined with the amylograph have been demonstrated. described {Jfi), In general, amylograph curves of Some of the chemical and physical tests are the rice varieties were typical of those of other cereal determination of amylose content {107), starch- starches, but they showed appreciable differences iodine-blue value (^i), gelatinization tempera- among the varieties studied. Long-grain varieties ture {Ifi), type and extent of disintegration of with the highest amylose content usually showed whole milled kernels in contact with dilute alkali the greatest increases in viscosity when cooled to {79), and amylograph pasting qualities {Jfi)^ 50° C. Amylograms of most short- and medium- The amylose content of rice, particularly of grain varieties generally exhibited relatively long-grain types, has recently been associated shorter gelatinization times. with cooking quality (5^, 107), The investigations Notable varietal exceptions within each grain of Williams and others {107), showed that the type were observed by Halick and Kelly {J^O), long-grain domestic varieties known to cook dry For example, the long-grain varieties Rexark and and ñaky usually had the highest amylose con- Toro had amylose contents, gelatinization tem- tent; whereas the amylose contents of the short- peratures, and alkali spreading and clearing reac- and medium-grain varieties investigated were tions similar to those of typical short- and medium- somewhat lower. The glutinous (waxy) varieties grain varieties. These long-grain varieties are also contain virtually no amylose. The simple, rapid, thought to resemble the typical short- and me- and somewhat empirical starch-iodine-blue test dium-grain varieties more in cooking quality than {il) is particularly useful in the breeding pro- they resemble other long-grain varieties. Of all the grams for estimating the relative amylose content varieties tested, Century Patna 231, a long-grain of early-generation breeding material. variety, and Early Prolific, a medium-grain vari- The gelatinization temperature of rice is be- ety, had the highest gelatinization temperature lieved to be closely related to cooking quality. and were the most resistant to the action of dilute The amylograph studies {Jfi) showed that most alkali. In general, these varieties have not been short- and medium-grain varieties gelatinized at widely accepted for certain types of cooked rice lower temperatures than did most of the long- and processed products. grain varieties investigated. These results were Average values for some of the physical and confirmed by granule swelling birefringence end- chemical characters of 18 rice varieties grown in point temperature (BEPT) determinations {S9), the Uniform Yield Nurseries in Arkansas, Louisi- The reaction of milled rice kernels in contact ana, Mississippi, and Texas, from 1960 through with dilute alkali has been used to classify the 1969, are tabulated in table 8. Environmental and cooking quality of rice {70, 100), The type and other factors influence these qualities to some ex- extent of disintegration of whole milled rice ker- tent; however, within a limited range, the values 44 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE are representative of each variety. In the rice- and the time used in performing the tests. Par- breeding program, characters of new selections are boiling and canning tests using 5 to 10 grams of always compared with those of comparably grown rough rice can be conducted on rather large num- commercial varieties. Some of the physical and bers of samples if necessary, but the cost is rela- chemical characters of the typical cooking long- tively high. grain varieties in the United States and listed in The cross Gulf rose X Bluebonnet 50 is an ex- table 8 are : A relatively high amylose content ; low ample of how quality tests aid in the rice-breeding starch-iodine-blue value ; an intermediate gelatini- program. In this cross, the objective was to de- zation temperature, a slight-to-moderat?e reaction velop a long-grain Bluebonnet 50 type possessing of head rice in contact with dilute alkali (Alkali the hoja blanca resistance of Gulf rose. Gulf rose spreading value) ; and a moderate water uptake ordinarily shows a low gelatinization temperature capacity at 77° C. Characters of the typical cook- and relatively low amylose content, whereas Blue- ing short- and medium-grain varieties in the bonnet 50 ordinarily shows intermediate gelatini- United States are : A relatively low amylose con- zation temperature and relatively high amylose tent; high starch-iodine-blue value; a relatively content. low gelatinization temperature ; a pronounced and A large number of long-grain types resembling extensive reaction of head rice in contact with di- Bluebonnet 50 were saved from the F2 plant lute alkali; and a relatively high water uptake population. A part of the grain from each plant capacity at 77° C. Parboil-canning stability of the saved was milled in a test-tube miller. The milled typical long-grain varieties in terms of percent samples were visually examined for grain texture, solids lost during canning is relatively low, and size, and shape; and alkali digestion and iodine- the canned kernels show little splitting and fray- blue values were determined. Samples showing ing of edges and ends. Parboil-canning stability of grains of intermediate gelatinization temperature the typical short- and medium-grain varieties in and relatively high amylose (an iodine-blue value terms of percent solids lost during canning is rela- of 25 or below) were saved for hoja blanca testing. tively high, and the canned kernels show extensive In the F3 plant generation, a bulk made up of 15 splitting and fraying of edges and ends. or more panicles was again milled, and the quality Physical and chemical grain quality evaluation tests were repeated. The strains that appeared tests are invaluable to the plant breeder in develop- satisfactory agronomically after F5 or ¥& plant ing varieties with specific processing and cooking generations were increased for yield testing; and quality. The specific tests used depend on the qual- at this stage other quality tests for characters, such ity variables of the particular hybrid population as analytical amylose and protein content and past- in question and the purpose for which the end ing viscosity, were made. product is to be used. The results of the starch-iodine-blue and the Breeding for Disease Resistance alkali digestion tests enable the plant breeder to classify early-generation hybrid lines as to gelat- The easiest, most practical, and least expensive inization temperature of starch and relative way to control rice diseases is to use resistant vari- amount of amylose. Promising advanced-genera- eties. Considerable research has been done in the tion lines are evaluated from data obtained in United States on the reaction of rice varieties to analytical tests for amylose and protein content, the various rice diseases and on developing resist- pasting qualities using an amylograph, and cook- ant varieties. Much of this work has been reviewed ing, parboiling, and canning tests. In certain (iJ, 16). The symptoms, control measures, and crosses where it is essential to retain a specific importance of the rice diseases occurring in the pasting quality or parboiling and canning char- United States are given in the section "Kice Dis- acter, it may be necessary to obtain amylograph eases,'- p. 141. Methods used to breed resistant vari- data on early-generation lines to make certain that eties are presented in this section. the desired qualities are recovered. Such tests are Certain basic principles apply to breeding for usually conducted on a relatively small number of disease-resistant varieties of a crop. Techniques to samples because of the quantities of grain required create an epiphytotic of the disease and to evaluate RICE IN THE UNITED STATEiS 45

0 (M CO CO CO CD l> Tí< Ttl CO CO CO CO CO CO CO CO CO (N^(MCq(MT-H(McO'-H 00 ■4 bO >s i i>. rO '^k') 5^5 ÛH

5S O rH 00 05 CO 10 (M -^ '—I 0»OC005COt>(Mr}HC^ 10 LQ CO CO CO CO CO ï>CÔC0^C0 1>-'cûCÛI>'

0 (MTÍHCOOOCO(MCOCOlr^ si CX) CX) 00 C5 0 05 05 C75 00 0000001>OOCOI>OOÍ^ •^ c^ rOi ? 5S cq ?5

CD LO 00 T-H t>- CO i-H CO CO iOlOO500(MC0COÍ>'^ LO I> O Tt^ CO (M CO COCOCO'<^(MC0C0I-HC^ 0 CO CO CO CO CO CO CO CO CO rHrHr-lrHrHrHT-lTtlTTH ^ S

è§ rö ^ 0000000 0000 O O T3 T5 'T3 T3 X5 T3 T3 T3 TJ 'X3 T3 T3 13

o Ö 9 tí I—I 1 I I I I 1 HH I—I

l> C^ LO '^ TJH ^^ CO 050LO(MOcoa5i>c^ CO CO CO CD CO CO CO CO TÍÍ TÍH '^' TÍH T^' CO CD Tfi

tpi

Ä, O Ä, tM c^ Tl üO g 00 O 10 1> T-H CO '^ "^Tt^t^Ti^t^CO^OCD 1 CO C^4 Ttl T^l UO LO LQ CO CO ^ 0 o> bO 1 í3 'S eo o3 03 "-HN M g OJ > 1$ O3T3 '^ CO > -M n o3 CÄ ^ 3 \§

CT) b- C5 00 00 I> IOCO^I>^COCOCÏIO Oí -(j s^ 00 0 tí s ^ (M

s 0

ioioc<ï05ooi>oo O5xa5oot^cocoo5co T3tí 03 I o3 tí ^ c3 ' tí O I-:;

o5 'S a bß O tí O 'S o3 o Ö 0 .9 CO 05 í w bß g -^ tí c3 o o a n ^ -^ 2 ^ ^ bJÜ a ^ u o .2 <1 O O ^ ^ œ cS a:sa;â^pqpqpqPQtío^eHH ;-! T3 .tí0 ZQ 488-871 O- 46 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICUI/TURE

and record the response of the plant to the causal (1) Seeds from F3 or F4 progenies of crosses organism must be developed. Then the available between varieties possessing the desired resistance varieties and breeding lines must be tested to genes are saved. determine sources of resistance to the causal organ- (2) Seedlings from these progenies are inocu- ism. Eesistant varieties can then be crossed with lated in the greenhouse to determine reaction to varieties possessing other desirable characters, the race 6. Susceptible selections are not tested further. progenies can be tested, and the resistant lines can (3) Seedlings that are resistant to race 6 are be isolated. Where a resistant variety has few then inoculated in the greenhouse with race 1. attributes other than resistance to the disease under (4) Selections that carry resistance to races 1 study, it may be used as the donor variety in a and 6 are sown in the field for further evaluation backcross program. In order to develop efficient of field and grain characters. Many of the less de- breeding techniques, the mode of inheritance of sirable lines are eliminated at this stage. reaction to causal organisms must be determined. (5) Seedlings of the selected lines are inoculated in the greenhouse with races 1 and 6 to confirm Blast results of previous tests. (6) Selections that are resistant to races 1 and Blast, caused by the fungus Pyricularia oryzae 6 then are inoculated with races 2 or 4, 7, 8,16, and Cav., is an important disease of rice. Much re- 19, as time and facilities permit. search has been done on it in the United States (7) Selections that are resistant to all races are and in other rice-producing countries. Physio- then grown in the field in observation rows or in logical races of P, oryzae that occur in the United yield tests for further evaluation as potential new States {78) make the breeding of resistant varie- rice varieties. ties more complex than if races did not occur. Selections that are resistant to the desired races Genes for resistance to each race that is known to of blast but do not appear promising for develop- occur in the United States are available, but not ing an improved commercial variety are crossed all are present in any one variety. with a suitable variety or strains. Testing for reac- A satisfactory method for inoculating plants tion to blast is then repeated. This operation may {10) and a system of rating reaction have been be repeated several times, or until suitable plant developed {78). Many varieties in the World Col- types occur. lection and breeding lines have been tested under Many breeding lines have been tested, and prom- controlled conditions for reaction to specific races ising strains of all grain types have been devel- of P, oryzae. Uniform trials have been conducted oped to combine resistance to several of the more in the field under a wide range of environmental common races of blast found in the United States. conditions {15), Some studies on the genetics of In greenhouse tests Dawn (^^), Nova (5<^), and reaction to P, oryzae have been made (i<5, 81), but Saturn {51) were resistant or moderately resistant information is not available on the genetics of to most of the races of blast. In field tests these reaction to all races. varieties were considerably more resistant than An accelerated program of testing and breeding were other varieties. Since it was released in 1966, rice for resistance to blast in the United States Nova 66 {61) has replaced Nova. was started in 1959. The initial tests consisted largely of screening in the greenhouse the more Brown Leaf Spot promising varieties and selections for reaction to Brown leaf spot, caused by the fungus Helmin- races 1 and 6. Later, many other breeding lines thosparium oryzae B. de Haan, is a common dis- from Arkansas, Louisiana, and Texas rice experi- ease of rice in humid areas. In 1941 a report was ment stations were screened. Material grown in the published {5) concerning the mode of inheritance field and harvested in the late summer or early of resistance to Hehrdnthosporium in a cross of a fall is available for greenhouse testing from No- moderately resistant and a susceptible variety. vember through March. Inoculation with conidia was used to induce infec- The method used for developing blast-resistant tion. All gradation from moderately resistant to varieties in the United States {15) is as follows: susceptible occurred in the segregating popula- RICE IN THE UNITED STATEiS 47 tions. This indicated that the reaction was con- of infection. In Louisiana, a number of resistant trolled by several genes lacking dominance. selections were obtained when Blue Rose was The reaction of seedlings grown in the green- crossed or backcrossed to Rexoro, although both house showed a fairly close relationship with that parental varieties were susceptible to one or more of mature plants grown in the field and suggested races. the possibility of early elimination of a large pro- portion of susceptible segregates in breeding for Stralghthead resistance (5), According to Nagai (5/), he and Hara studied Straighthead is a physiological disease that occurs under certain environmental conditions in an especially susceptible mutant, and susceptibility the United States. A breeding program to develop proved to be due to a single recessive gene in crosses of a very susceptible and a normal plant. straighthead-resistant varieties was begun in the United States in 1953, at Eagle Lake, Tex. The A breeding program to develop varieties resist- method of testing consists simply of drill seeding ant to brown leaf spot was begun at Beaumont, the test entries in an area with a soil type such as Tex., about 1938. C.I. 9515, a resistant but agro- Hockley fine sandy loam, conducive to straight- nomically undesirable strain selected from a cross head development and keeping the nursery test made at Beaumont in 1938, was crossed with popu- area continuously submerged after the initial irri- lar long-grain commercial varieties in 1945. Resist- gation (99), The reactions of the American rice ant selections from the 1945 crosses were crossed to varieties (IS, H) served as a basis for selecting commercial long-grain varieties in 1959. Selecting parental varieties for resistance in a breeding pro- for resistance was started in the F3 population by saving the more resistant plants in rows showing gram. Straighthead resistance is relative, and none of the varieties tested are immune or highly re- the best resistance, along with other desirable sistant. The resistance of Bluebonnet, Bluebonnet quality. A highly susceptible spreader variety was 50, Lacrosse, Prelude, and Toro is derived directly sown adjacent to each selection. Differences in dis- or indirectly from Fortuna, selected from the Pa ease reaction were relative and were more apparent Chiam variety obtained from Taiwan. CI. 5094, between panicle rows than among individual selected from the Sinanpagh variety from the plants. Resistance was based on the number and Philippines, served as the source of resistance in size of spots. The studies indicated that inheritance Texas Patna and the original Century Patna. is not simple and probably involves several genes. Testing selections from several crosses was begun Selecting for resistance was continued through in 1954. For example, straighthead-resistant selec- several generations or until apparent true breeding tions were recovered from a cross of Bluebonnet lines were established. Resistant lines possessing good agrnomic character were selected from the (resistant) X Century Patna 231 (susceptible) 1959 crosses. and from backcrosses to Century Patna 231. Al- though a number of the straighthead-resistant se- Narrow Brown Leaf Spot lections from the Century Patna 231 crosses were promising and were seriously considered as new Narrow brown leaf spot, caused by the fungus varieties, none were released. Belle Patna, from a Cercospora oryzae I. Miyake, is a common disease different cross, was released at a new variety in of rice in the Southern States. There are physio- logical races of the causal fungus {SS^ 89-91)^ but 1961 {25). This variety has Bluebonnet as its genes for resistance for each race are available. source of straighthead resistance. The mode of inheritance of reaction to C, oryzae In the breeding program, straighthead-resistant has been studied (J, 52^ 5S), No linkage between selections have been obtained readily by selecting genes for reaction to Cercospora and those for ex- panicles from resistant or segregating lines for pression of five qualitative characters was found retesting in panicle rows. The number of genes (52), Most of the work on the development of involved in straighthead reaction was not estab- Oercospora-vesist^iit varieties has been done in the lished, but inheritance appeared to be relatively field under conditions of natural infection. Selec- simple in crosses within the long-grain groups of tion is based on type of lesion and relative severity varieties. Resistance seems to be dominant (i^). 48 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

White Tip quent tests. Nova, released in 1963, also a hoja blanca-resistant variety {58), was an increase White tip, first observed in Louisiana before from a selection rated as resistant in early tests. 1930, was considered to be a physiological disease Testing and breeding for resistance to hoja until 1949, when it was found to be caused by a blanca are being continued in cooperation with seedborne, foliar nematode, Aphelenchoides ies- Government and private agencies in several Cen- seyi Christie {SJf.), However, by that time consid- tral and South American countries. Much of the erable progress had been made in selecting for testing of varieties and breeding lines for reac- resistance in nurseries in which the disease oc- tion to hoja blanca has been done under conditions curred rather consistently each year. of natural infection in the field. This work has After white tip was found to be caused by a seed- been done in foreign countries where epiphytotics borne nematode, various methods were used to of the disease usually occur. Hybrid and back- insure infection in order to determine the reaction cross plants and progeny lines also are tested in of varieties and selections to white tip. These the greenhouse where plants are inoculated with methods consisted of including heavily infested viruliferous vectors. Since resistance to hoja seed of susceptible varieties as spreader rows, blanca is dominant, backcross plants that carry using rice hulls containing large numbers of nema- resistance can be identified in greenhouse tests. todes, or introducing nematodes from laboratory Backcross seeds are planted in the greenhouse or cultures into the irrigation water. Atkins and field. As soon as the plants tiller, they are di- Todd {17) determined reaction of many rice vari- vided and part of the plant is tested for reaction eties in the united States to white tip. No studies to hoja blanca. The reaction can be determined have been made on the genetics of inheritance. by the time the plant flowers so that plants carry- Hoja Blanca ing resistance can be identified and used as par- ents in the backcross program. Hoja blanca is a virus disease of rice that oc- The pedigree method, the backcross method, curs in the Western Hemisphere. In 1957, a large and a modification of these two methods were number of United States varieties and selections, used in Texas to develop long-grain varieties as well as introduced varieties, were tested for resistant to hoja blanca. The material used in hoja blanca reaction under conditions of natural the pedigree method was obtained by crossing the infection in Cuba and Venezuela {12), On the long-grain variety Bluebonnet 50 and several basis of marked differences in disease reaction in promising long-grain hybrid selections with the nursery tests, Arkrose, Asahi, Colusa, La- the hoja blanca-resistant varieties Gulfrose and crosse, Mo. E-500, and several experimental va- Tainan-iku No. 487 (P.I. 215,936). The Fi rieties were designated as sources of hoja blanca plants were grown in the greenhouse during the resistance for use in breeding. winter of 1957-58, and the F2 populations were Since both resistant and susceptible entries were grown in the field during the summer of 1958. found among a number of advanced-generation In the fall of 1958, the seeds from 353 carefully selections from two crosses having Lacrosse as one selected F2 plants were saved. A few grams of parent, it was concluded that the genes for resis- grain from each selection were milled in a test- tance could be readily transferred in crosses. Other tube miller, and alkali digestion and iodine-blue studies showed that resistance was genetically con- tests were performed. In 1959, progenies from trolled and could be transferred to varieties of all each F2 plant were grown in Colombia, Cuba, grain types {2Jf) and that resistance was dominant and Venezuela to test their reaction to ho jo {23). Since 1957, other lines resistant to hoja blan- blanca ; at Eagle Lake, Tex., to test their reaction ca have been recovered from several crosses be- to straighthead ; and in the nursery at Beaumont, tween suscepitible and resistant parents. Thus far, Tex., to advance each selection on a pedigree basis none have been released as varieties. and to select on the basis of plant type and vigor. Gulfrose, released in 1960 as a hoja blanca- In 1960, scientists at Beaumont selected 507 resistant variety (7, 35), was an increase of a panicles from the better F3 lines grown in 1959. selection rated as resistant in the 1957 and subse- Seed from each of these selections was sent to RICE IN THE UNITED STATEIS 49 Colombia for further hoja blanca testing and ties were tested for yield and other field characters progeny of each was grown in the breeding nurs- at Beaumont in 1963. ery at Beaumont. About 20 panicles were har- Since hoja blanca resistance is dominant, Blue- vested from each row, and this grain was used to bonnet 50 was used as the recurrent parent in the determine the cooking and processing quality of modified backcross method. The hoja blanca re- the F4 lines. action of Fi backcross plants was obtained at the Excellent hoja blanca readings from Colombia hoja blanca testing laboratory at Baton Rouge, were obtained on the F4 lines tested in 1960, as La. In 1963, Fi plants with four backcrosses to well as the F5 and ¥& lines tested in 1961 and Bluebonnet 50 were tested for hoja blanca reac- 1962. Hoja blanca-resistant selections were tested tion. Some of these plants are similar to Blue- further for reaction to straighthead at Eagle bonnet 50 in plant and grain type and they are Lake, for reaction to blast of seedlings at Beau- resistant to hoja blanca. mont, and for quality behavior in the cooperative Eice Quality Laboratory at Beaumont. Description of Varieties In 1962, a number of the more promising strains carrying hoja blanca resistance were grown in The names and accession numbers of the 18 advanced yield trials at Beaumont. From the varieties reported in this section of the hand- crosses made in 1957, hybrid selections of early book are listed in table 9. These include the prin- and midseason maturity and short-, medium-, and cipal commercial varieties in the United States. long-grain types that are promising for resistance to hoja blanca are now available for extensive Short-Grain Varieties testing. In 1970, about 8.8 percent of the total rice acre- In the modification of the backcross method age in the united States was sown to short-grain used in the hoja blanca breeding program, the (Pearl) varieties {88). In Arkansas, Louisiana, Fi plant of the cross Bluebonnet 50 X Gulfrose Mississippi, and Texas, only 0.30 percent of the was backcrossed to Bluebonnet 50 in 1958. A total crop produced was the short-grain type, whereas of 31 backcrossed seeds produced plants. The seeds in California, 46.8 percent was this type. from each plant were sent to the hoja blanca test- The commercial short-grain varieties grown in ing laboratory, then located at Camaguey, Cuba, yield tests were Caloro and Colusa. Rough rice and backcross populations carrying hoja blanca and milled kernels of each of these two varieties resistance were identified. Space planted popula- are shown in figure 21. These and other short- tions of each backcross line also were grown in grain varieties have rather slender culms, narrow the breeding nursery at Beaumont in 1959, and (about %-inch) leaf blades, and yellow or straw- plants were selected from the lines that were colored rough hulls enclosing the kernels. A plant promising for hoja blanca resistance. In 1960, F3 of Caloro is shown in figure 22. When milled, lines were tested for hoja blanca reaction in Co- short-grain varieties usually yield as high a per- lombia and for straighthead reaction at Eagle centage of head rice (whole grain milled) as do Lake. F2 lines were also sown at Campo Cotaxtla, the medium-grain varieties and higher than do the near Veracruz, Mexico, in the fall of 1959; and long-grain varieties. Short-grain varieties are not two crops per year were obtained through 1961. grown in the Southern States because the varieties available are not well adapted for this area and Selections from the Mexico nursery were sent to there is little demand in this area for rice of this Colombia for testing for hoja blanca reaction and type. to Beaumont for grain quality testing. By the CALORO.—Caloro was selected from Early end of 1961, many long-grain selections promising Wataribune in 1913 at the Biggs Rice Field Sta- for hoja blanca resistance were available for ex- tion, Biggs, Calif., and was distributed in the tensive testing at Beaumont. Long-grain selections spring of 1921. It is an early to midseason, partly of early and midseason maturity that are resistant awned variety that heads and matures uniformly to hoja blanca and straighthead and that possess in California and produces relatively high yields Bluebonnet milling, processing, and cooking quali- with reasonably good milling quality on either 50 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

TABLE 9.—Name, C.I. number, FAO number, and registration number jor reported rice varieties

Grain type and variety C.I. FAO Registration number ■ number ^ number ^

Short-grain : Caloro 1561-1 211 5 Colusa- 1600 213 8 Medium-grain : Arkrose 8310 207 1 Calrose 8988 1013 6 CS-M3 9675 34 Nato 8998 1133 13 Nova 66 9481 30 Saturn 9540 1415 29 Zenith 7787 206 19 Long-grain : Belle Patna 9433 1334 27 Bluebelle 9544 32 Bluebonnet 50___ 8990 1012 3 Dawn 9534 33 Delia (aromatic)- 9483 Rexoro 1779 214 14 Starbonnet 9584 31 Toro 9013 1134 18 TP 49 8991 1020 17

^ Accession number, Agricultural Research Service. 2 Accession number, World Catalogue of Genetic Stocks- -Rice. Food and Agriculture Organiza- tion of the United Nations. 3 Registration number, American Society of Agronomy- -U.S. Department of Agriculture.

virgin soil or old ricelands. It is the leading short- in Missouri and fairly well on fertile land in Ar- grain variety grown in California, yields well in kansas, but it frequently lodges when the crop is Arkansas {59,60^82) and Missouri (77), and yields heavy. Colusa has been for many years, and still reasonably well in Louisiana and Texas. Although is the most popular early-maturing variety grown it appears to be adapted for growing under a wide in California. range of conditions, it is grown principally in Cali- Medium-Grain Varieties fornia. Caloro is the most important short-grain variety grown in the United States. In 1970, about 41.5 percent of the total rice acre- CoLusA.—Colusa {32) was selected in 1911 at age in the United States was sown to medium- the Rice Experiment Station, Crowley, La., from grain varieties {88), In the Southern States about Chinese, a variety introduced from Italy in 1909 38.9 percent of the acreage was sown to this type by Haven Metcalf. It was tested at the Biggs Eice and in California about 53.2 percent. Of the total Field Station, Biggs, Calif., and distributed in rice sown in the United States in 1970, Saturn 1917 and 1918. Colusa is an early-maturing, awn- made up 16.9 percent; Nato, 13.2 percent; Calrose, less variety of reasonably good milling quality that 9.8 percent; Nova, 1.7 percent; and other medium- heads and matures rather uniformly and that pro- grain varieties such as Arkrose, less than 1 per- duces relatively high yields on fertile land. In Cal- cent. Calrose was grown only in California and ifornia it is much less productive than Caloro on the other medium-grain varieties were grown only old riceland of average fertility. However, when in the Southern States. nitrogen fertilizer is applied at fairly high rates, The principal medium-grain varieties grown in it produces as much as Caloro. Colusa yields well the yield tests were Arkrose, Calrose, CS-M3, RICE IN THE TINITED STATES

BN-22039 FIGURE 21.—Rough rice and milled kernels of (A) Caloro and (B) Colusa.

Nato, Nova 66, Saturn, and Zenith. Kough rice a partly awned variety, very similar in growth and milled kernels of each of these seven varieties habit and maturity to Caloro. In California, Cal- are shown in figure 23. These varieties, with the rose appears to be equal to Caloro in yielding exception of Calrose, have rather stout culms and capacity and in milling quality. It stands up well, relatively wide (about %-inch) leaf blades. A matures evenly, and is as easy to combine as plant of Nato is shown in figure 22. When milled, Caloro. Calrose was grown on a small acreage for medium-grain varieties usually yield more head the first time in 1948, and in 1970 it was grown on rice (whole kernels) than the long-grain varieties 53.2 percent of the California rice acreage. yield. CS-M3.—CS-M3 (80) is a medium-grain vari- ARKROSE.—Arkrose (71) was selected at the Rice ety selected at Briggs, Calif., from the cross C6 Branch Experiment Station, Stuttgart, Ark., from Smooth X Calrose. It is a sparsely awned variety, the cross Caloro X Blue Rose. Arkrose was dis- with glabrous lemma, palea, and leaves. It pro- tributed in 1942. Arkrose matures about a week duces average grain yields, somewhat higher than earlier than Supreme Blue Rose, and in some sec- those of Calrose in California. CS-M3 is similar to tions of Arkansas it is meeting the demand for a Calrose in kernel shape and size, plant height, date Blue Rose type. Arkrose yields well and is rela- of heading and maturity, and such quality factors tively easy to thresh. It is similar to Blue Rose in as iodine-blue and alkali-digestion values. CS-M3 milling and table quality. It is more difficult to is equal or superior to Calrose in translucency of dry artificially than are most long-grain vaiieties. kernels and in milling characteristics. It is adapted Arkrose has been grown in Arkansas and, rarely, to those ai'eas in California where Calrose is grown. in Texas. NATO.—Nato was selected from the cross CALROSE.—Calrose (71) was selected at the (Rexoro X Purple-leaf) X Magnolia at the Rice Biggs Rice Field Station, Biggs, Calif., from the Experiment Station, Crowley, La. It was released cross Caloro X Calady backcrossed to Caloro. It is in 1956. It is an early-maturing variety with com- 52 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

BN-22013 BN-22012 FIGURE22.—Plants of (A) Caloro, (B) Nato, and (0) Rexoro.

paratively short, strong straw; it produces good mum head rice (milling) yields. Kernels of Nova field and mill yields ; and it is suitable for making 66 are slightly larger than those of Nato but are dry cereals and for other uses for which varieties very similar in processing and cooking character- of similar type are adapted (4^). Nato was the istics. leading rice variety in the United States for sev- SATURIST.—Saturn (51) is an early variety se- eral years. lected at the Rice Experiment Station, Crowley, NOVA 66.—Nova 66 (61, 63, 94) originated as a La., from the cross Lacrosse X Magnolia and re- single plant selection from the variety Nova. The leased in 1964. It is the most productive variety latter was selected from the cross Lacrosse X grown in southwest Louisiana but is subject to Zenith-Nira and released from the Rice Branch considerable losses from lodging under adverse Experiment Station, Stuttgart, Ark., in 1963. Seed weather conditions. The grain is similar to Nato of Nova 66 was distributed from the same station in milling and quality. Saturn is resistant to one in 1966. Nova 66 is a short-season (early-matur- or more races of blast to which Nato is susceptible. ing) variety with moderately strong straw and has It is resistant to kernel smut, but susceptible to the potential for producing high grain yields. Use stem rot. In 1967 it became the leading variety in of the proper rate and timing of nitrogen fertilizer Louisiana and in 1970 was grown on 54 percent applications is very important in order to obtain of the acreage. the full benefit from the lodging resistance and ZEXITII.—Zenith (71) was selected from Blue high yielding potential of Nova 66. When con- Rose in 1930 by Glen K. Alter, near DeWitt, Ark. ditions are favorable for severe lodging, plants of In 1931, several selections were tested in the co- Nova 66 tend to bend over about 8 to 10 inches operative breeding program at the Rice Branch above the ground in contrast to plants of Nato Experiment Station, Stuttgart, Ark. Of these and certain other varieties that usually lodge at selections, Arkansas 141-8 proved to be the best. the soil surface. Nova 66 requires more care in It was named Zenith and distributed in 1936. handling and drying than Nato to obtain maxi- Zenith is an early-maturing, awnless variety, and RICE IN THE UNITE'D STATES 53 it is uniform in heading and in maturing. In these varieties have rather large culms and rela- 1954, Zenith Avas grown on over 50 percent of the tively wide (about s/g-inch) leaf blades. A Rexoro rice acreage in the Southern States. It has since plant is shown in figure 22. Because of the long been replaced by the shorter strawed, smooth- growing season required, Rexoro and TP 49 are hulled Nato variety. Strains similar to Zenith grown only in Louisiana and Texas. have been isolated from Early Prolific. BELLE PATNA.—Belle Patna was selected from the cross Rexoro X (Hill selection X Bluebonnet) Long-Grain Varieties at the Agricultural Research and Extension Cen- In 1970, about 49.7 percent of the total rice ter and was released in 1961 (2). It is a very acreage in the United States was sown to long- early-maturing, slender-grain variety with cook- grain varieties (88). Of the total rice sown in the ing quality similar to Rexoro. Southern States, Starbonnet made up 20.8 percent; BLUEBELLE.—Bluebelle {£8, 29) was selected Bluebelle, 14.9 percent; Belle Patna, 9.3 percent; at the Agricultural Research and Extension Center Bluebonnet 50, 2.8 percent; Dawn, 1.5 percent; from the cross C.I. 9214 X (Century Patna 231 and Rexoro, TP 49, and Toro, each made up less X C.I. 9122). C.I. 9214 is a rogue from Rexark, than 1 percent. and C.I. 9122 was from the cross Long-grain Hill The principal long-grain varieties grown in the X Bluebonnet. Bluebelle was released to farmers yield tests were Belle Patna, Bluebelle, Bluebonnet in 1965. It matures a few days later than Belle 50, Dawn, Delia, Rexoro, Starbonnet, Toro, and Patna but it is classed as a very early maturing TP 49. Rough rice and milled kernels of each of variety. Bluebelle grain is similar to Bluebonnet these eight varieties are shown in figure 24. Most of 50 in size, shape, and quality. It is superior to

:.—Rough rice and kernels of (A) , (B) Calrose, CS-M3, (D) Nato, Nova 66, (P) Saturn, ) Zenith. 54 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE.

Blueboimet 50 and Belle Patna in lodging resist- Bluebonnet 50 is a midseason variety. It has ance and grain yield. relatively short, stiff straw, and the grains are BLTJEBONNET 50.—Bluebonnet 50 was selected partly awned. It yields and mills well and has from Bluebonnet in 1950 at the Agricultural good cooking quality, being comparable to Rexoro Eesearch and Extension Center. Bluebonnet and Texas Patna in those characters. The grain was a progeny of the cross Eexoro X Fortuna. of Bluebonnet 50 is thicker than that of Rexoro Seed of Bluebonnet 50 was distributed in 1951. but is more slender than that of Nira. Bluebon-

FIGURE 24.—Rough rice and milled kernels of (A) Belle Patna, (B) Blue- belle, (C) Bluebonnet 50, (D) Dawn, (fí) Delia, (F) Rexoro, ((Ï) Star- bonnet, (íT) Toro, and (7) TP 49. RICE IN THE UNITED STATEiS 55

net 50 is well suited for harvesting by the combine- those of Bluebonnet 50. Seed of Starbonnet was drier method. distributed from the Rice Branch Experiment DAWN.—Dawn {26, 27, 31, 56, 63) was selected Station in 1967. Because of its high field and mil- from the cross Century Patna 231 X (TP 49 X ling yields and lodging resistance, it rapidly re- C.I. 9515) at the Agricultural Eesearch and Ex- placed Bluebonnet 50. In 1969, the third year it tension Center, Beaumont, Tex. It was released to was grown commercially, Starbonnet led all other farmers in 1966. Dawn is about 2 weeks earlier and varieties in the United States in total production, it is slightly shorter in height than Bluebonnet 50. and in 1970 it led in total acreage. Dawn is resistant to the pathogenic races of rice ToRO.—Toro was selected from a cross in which blast in the Southern States. the varieties Bluebonnet, Blue Rose, and Rexoro were the parents. It was developed at the Rice DELLA.—Delia is an aromatic variety, having an Experiment Station, Crowley, La., and released odor and flavor similar to that of popcorn. It is in 1955 {Jf8). It is similar to Bluebonnet 50 in about 1 week later than Saturn, fairly tall, and plant height and straw strength. It yields and moderately productive. It mills well and the milled mills well, but it is about as hard to thresh as rice is attractive in appearance. Della was selected Zenith. The grain type is similar to that of Blue- at the Rice Experiment Station, Crowley, La., bonnet, but the cooking quality of these two va- from the cross (Rexoro X Zenith selection) X rieties is different. Toro, when cooked, is firmer R-D, a late-maturing selection from the cross Rex- than Bluebonnet but not as dry and flaky. oro X Delitus, which received the flavor from the TP 49.—TP 49 {57) was selected from the cross Delitus parent. Della was released as a specialty Texas Patna X (Rexoro X C.I. 7689) at the Agri- variety in 1971. cultural Research and Extension Center and re- REXORO.—Rexoro {32) was selected in 1926 at leased in 1951. It is a late-maturing, slender-grain the Rice Experiment Station, Crowley, La., from variety, similar to Texas Patna except that it has the Marong-Paroc variety introduced from the shorter and stronger straw and somewhat thicker Philippine Islands in 1911 by the U.S. Depart- grain. ment of Agriculture. Rexoro was distributed by the Department in cooperation with the Louisiana Other Kinds of Rice Agricultural Experiment Station in 1928. Rexoro Most of the rice varieties grown in the United is a stiff-strawed, late-maturing, slender-grain rice States are generally called common varieties ; that that yields and mills well for a variety of this is, they have no distinctive flavor or odor when type. The cooking quality is very good. cooked, and the starch in the endosperm contains STARBONNET.—Starbonnet {63, 67, lOJf) was both amylose and amylopectin. There are, how- selected from Fe generation of the cross Century ever, minor acreages of two other kinds of rice— Patna 231 (C.I. 8993) X Bluebonnet (CI. 8322), aromatic or scented, and glutinous or waxy. made at Stuttgart, Ark., in 1954. AROMATIC OR SCENTED.—Varieties having a dis- Compared with Bluebonnet 50 growing under tinctive odor and flavor somewhat like popcorn similar conditions, plants of Starbonnet produce when cooked are known as aromatic or scented more tillers, usually have shorter and narrower varieties ; they are cultivated widely in India and other southeast Asian countries. Scented rices are leaves, average 8 days earlier in heading, produce said to be low yielding; but because they are panicles that are more compact and less drooping, greatly esteemed, they may sell at twice the price have culms (stems) with considerably shorter of other rices of fine quality. The odor and flavor basal internodes at maturity so they average 15 are assumed to be due to an aromatic substance. percent shorter in height, and are much more re- Since aroma is sometimes noted in the growing sistant to lodging. Starbonnet grains usually are crop, it is not limited to the endosperm. The aroma awnless but under conditions highly favorable for occurs in waxy as well as in ordinary types of rice. vegetative growth, short awns may be produced Two varieties, Delitus and Salvo, were selected in florets at the tips of the panicles. It has process- from introductions released in the United States. ing and cooking characteristics very similar to Delitus has been grown commercially, and at- 56 AGRICULTURE HANDBOOK NO. 28 9, U.S. DEPT. OF AGRICULTURE

tempts have been made to place it on the market. series grown in the southern rice area were A cross between Kexoro and Delitus produced Del- grouped according to grain type and length of rex and K-D, which fully retain the flavor. E-D growing season. The groups are: I, short- and is a productive, medium-late variety that should medium-grain early varieties ; II, long-grain early make possible profitable marketing of scented rice varieties ; III, long-grain midseason varieties ; IV, as a food specialty. Delia, an early-maturing short- and medium-grain midseason varieties; V, selection from the cross R-D X Rexoro-Zenith, has long-grain late varieties; and VI, medium- and improved milling quality and a more translucent long-grain very early varieties. There usually endosperm. were 14 entries in each group, many of them ex- GLUTINOUS OR WAXY.—Waxy varieties, com- perimental varieties. monly called glutinous, differ from common vari- The varietal experiments at each location were eties in that they contain only amylopectin starch on soil that was typical of large areas in the re- in the endosperm. Glutinous rice is grown on spective States. Crop rotation, seedbed prepara- about 1,000 acres in California each year. It is tion, time and rate of seeding, fertilization, and grown as a specialty crop, and the acreage needed weed control used for these experiments were to meet market demands has been small. Historic- practices that were thought to be optimum for ally, the principal use of glutinous rice has been each area. for preparing oriental ceremonial foods and con- Average grain yields shown in table 10 were fections. In some countries, glutinous rice is har- obtained from 1961 through 1969 at the following vested slightly green, is lightly parched before locations : Rice Branch Experiment Station, Stutt- milling, and is used as a breakfast food. Recent gart, Ark.; Rice Experiment Station, Crowley research has shown that glutinous rice flour may La.; Delta Branch Experiment Station, Stone- find a special use in the frozen food industry. The ville, Miss. ; Agricultural Research and Extension fiour, when made into foods for freezing, such as Center, Beaumont, Tex.; and Rice Experiment white sauce and desserts, resists syneresis (separa- Station, Biggs, Calif. The actual number of years tion or weeping) when thawed after freezing. each variety was grown is shown. The yield of Mochi Gomi is the variety of this type grown each variety is compared with the yield of one or two standard varieties that were grown through- in California. It is a short-grain, midseason va- out the entire period. riety. Compared with Caloro, it has about the In Arkansas, the short-grain variety Colusa same straw strength and is about 6 inches shorter. produced the highest average yield except those The glumes are less pubescent, and the grain of the medium-grain varieties Nova 6ß and Saturn. matures 5 to 10 days later. The waxy-white ker- All medium-grain varieties except Arkrose pro- nels are opaque. duced higher average yields than the long-grain varieties. Performance of Varieties In Louisiana, the short-grain varieties produced The performance of rice varieties is studied at lower average grain yields than the higher yield- several rice experiment stations in the southern ing medium- and long-grain varieties. The highest rice area and in California. The grain yields and yielding medium-grain varieties were Nova 66 and milling quality data reported for the southern rice Saturn. Yield for CS-M3 was higher, but CS-M3 area in this bulletin were obtained from the Uni- was grown only 2 years. Of the long-grain varie- ties. Dawn and Bluebelle produced high average form Rice Performance Nurseries from 1960 yields. The late-maturing, long-grain varieties through 1969. In these nurseries, varieties are Rexoro and TP 49 produced quite low yields. grown in randomized, quadruplicated, drill- In Mississippi, Caloro produced a slightly seeded plots. These nurseries were grown each year higher average yield than the other short-grain in Arkansas, Louisiana, Mississippi, and Texas. variety Colusa. Nova 66 and Saturn produced Yields reported for California were obtained from highest average yields of the medium-grain va- water-seeded nursery plots. rieties. The average yield of these two varieties Entries in the Uniform Rice Performance Nur- was higher than that of the long-grain varieties. RICE IN THE UNITED STATEIS 57

o 00o

CO ^ CO o '3 (M TjH ^ (M T-H LO TÍH o

o o I ^ bß

CD o o o T-H '-H Tt^ CO o O Oi rH TH (N CO O WCOaSrHOLOi-lOí^

o 00 rtH '^ '«^ CO o (M OOrHOSb-COOOCiOíM ^ tí lO C0cOTtCOCOLOT-(0 5^ X 2Í CO TÍH TJH" CO" CO" TíT LO' JO" TíH" ríT lo" TíT TíT -çjT (^T TtT TfH^ co"

CO CO (N O OÏ X o ocrioooosTtHcoO'* r-i rH r-t T—I

c o tí T—1 ^ ^ O o íO) TtH 00 TtH (M r^ (N 00 b- o t^ ^ o o T-i O 1—1 o 00 o Oi o O 00 o r-i T—1 ■<-i T—1 1—1 1—1 T—1 ;H f ^ ^ bO 2 od CO lO -?tH r^ lO Tí^ O CO lO o > TH 00 CSJ LO CO 00 1—1 00 CO o l> CO 1—1 o 00 lO o TíT »O Tt^" lO íO lO TÍH Tí^ ríH íO ^ "«^"^ CO TíH'^ »o -^ '^ ^ ^:> ;H o O CO 00 CO CO <>J o 05 00 o o 02 o 00 Oi CO o tt-H 1—1 O LO 4^ CJtí tí tí C5 lO CO T-H o 1—I TÍH coQocooascMcoríHas o o CJi 05 TH o rH 1-1 o aioasi-HOscoasosco o O fe tíO) bD w 2 05 ^ o 00 00 CO CO 00 Ir^ "^oorHoooot-t^asoo Ti 0) 00 ^ T-H LO 00 CO lO CO CO OOOOCMOI>CO'^COrf< tí >, o o CM o CO rfH as o LO THOOCMOTIHCOLOT-ICO a a; TtHTÍHLOTÍiTtHLOTJH TjH'^ÍIrtHLO'^CMTÍHThlCM rC 5^ ;3 +3 ■^ o tí CO CO CM o c:s 00 o oaioooos'-^coo'^ a; N 03 r/} 00 ^ o O 5^ ^ 2 13 O ?í rH CM 00 CO os l> LO CO LO Tfl 1—I LO '—I »o CM CO t>> <4H O^ o o o rH 1-1 o os o as o Oi o as O « T—I rH tí 03 T3 S-s 00 00 o ai 00 t> i> CM LO o -^ CO o TH CM LO CM Tí^ LO O) LO CO as CO rf< CO CM o o CM T-H 1> CM ^ CO o TÍH 1—I LO T-H 00 CM <3 o ^ lO LO LOLOCOLOCOCOLO LOLOLOLOLO "H Ci,

cooo cococMoa^ooo oasoooas

ffl S. « Tí í^ o. i 03 .B tí

bß o QJ CO o 153 s s p s p ^ •§•3 .2 5 u o O to 58 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

Of the long-grain varieties, Starbonnet and formance Nursery samples before milling was de- Dawn produced the highest yields. scribed in the section 'Testing for Milling, Cook- In Texas, the short-grain varieties produced ing, and Processing Qualities", p. 38. average yields somewhat lower than those of the The average yield of total milled rice varied medium- and long-grain varieties. Of the medium- from 68 percent for Dawn and Starbonnet to 74 grain varieties, Saturn and Nova 66 produced percent for Caloro and Colusa. Short-grain varie- highest average yields. Of the long-grain varie- ties produced slightly higher yields of total milled ties, Bluebelle and Dawn produced highest average rice than did medium- and long-grain varieties. yields. The medium-grain variety Nova 66 and the long- In California, Calrose produced slightly higher grain varieties Dawn and Starbonnet produced the average yields than Caloro or Colusa. The new lowest yields of total milled rice. variety CS-M3 produced slightly higher yields than Calrose. Choosing the Variety The average number of days from seeding to maturity, average plant height, estimated straw Several factors affect the choice of a variety. strength, hull pubescence, and hull color for the These are listed by Johnston, Cralley, and Henry 18 varieties included in this handbook are given {60) as market demand, satisfactory yielding in table 11. ability, location, proposed seeding date, soil fer- Head rice yields for the varieties listed in table 7 tility or anticipated fertilizer practices, relative are reported as average values and as individual maturity, susceptibility to diseases that may oc- minimum and maximum (range) values. Average cur, and seed supply. Growers of large acreages yields of head rice varied widely among the varie- may wish to sow tw^o or three varieties that differ ties, from 54 percent for Dawn to 69 percent for in date of maturity and grain type in order to Nato. In general, the average yield of head rice extend the harvesting period and to provide rice was highest for the short- and medium-grain vari- of different types for the market. eties and lowest for the long-grain varieties. Toro, A grower should carefully consider market re- a long-grain variety yielding 64 percent of head quirements in selecting varieties to grow. He or rice, was an exception. The milling quality of this his marketing agent may have difficulty in selling variety has been noted in earlier publications (^/, his rice if it is of poor quality, even though lo- If7). Individual minimum and maximum values of cally the variety may produce high yields. Also, head rice yields also varied widely. This variability growing a short- or medium-grain variety on a was evident not only among varieties but also limited basis in an established long-grain district, within the varieties. Individual minimum head for example, can pose real problems to the grower, rice yields ranged from 29 percent for Colusa to driei's, warehousemen, and millers to prevent mix- 63 percent for Nato, whereas individual maximum tures of the different types. yields ranged from 62 percent for TP 49 to 73 Seeding rice on highly fertile soil, such as percent for Colusa and Nato. In general, the per- newly cleared woodland, "new ground," or a field centages of hulls range from 18 to 22, and the per- used as a reservoir just before being seeded to centages of bran content range from 8 to 12 for rice, would necessitate the choice of a stiff- most U.S. varieties. strawed variety that would be less likely to lodge. Milling quality was determined each year for Even on old riceland low in natural fertility, if the varieties grown in the Uniform Rice Perform- a grower anticipates using heavy rates of nitro- ance Nurseries. Included in table 7, p. 42, are the gen fertilization, with part of it applied relatively average milling yields of head and total rice for late, such a practice may make it desirable to each of 18 varieties grown in the Uniform Rice choose a relatively early-maturing variety. Grow- Performance Nurseries in Arkansas, Louisiana, ers in the northern part of ricegrowing areas in Mississippi, and Texas from 1960 through 1969. the United States are limited to early-maturing Milling yields were determined according to the varieties because of the somewhat shorter growing modified procedure of Beachell and Halick {21). season and cooler night temperatures than those Treatment and preparation of Uniform Rice Per- prevailing farther south. It may be desirable to RICE IN THE UNITED STATEIS 59

^O OÔOOOOO TÍ ^ TÍ ^ TÍ ^ F-1 f-* 2 c3 2 c3 2 o3 ö2 gp QQQQPpq O ;H o í-i o ^ o zn o m Ô ixi

^-2 tí PHPH PHPHOíCZíCCCCPH m ui m m ui uixn

I

; bi) bJO Ö Ö tí P -1^ tí O 2 -^ O tí 3 5*. o S>S^ 5^

CO CD CO CO

I ge 03 X ^^ H "to ríH 00 (N O 00 CO TíH l>C0OI:^I>TÍHrt^O(N Tí lO Tt^ LO lO -rí^ rt^ >0 TÍHT^lO'«tl'^LO^»OLO tí

o o CO O T^ lO (N Tíí 00 TH 00 1> O (N O l> "^ Tí^ LO '^ TÍH rti Tí^ Ttí CO "^ "^ Ttl i a 03 CO a ce o a i-H Til 1—I lO T^ Ttl 00 (NO00C0L0aiC01>05 a o ^g LO Ttl LO TjH -^ 'ííí TjH r^ ^^ ^^ TJH T^H »^ ^H '^H T^ >

1> í^ '^ 00 CO CO (M Tí^ o Oi CO CO o 00 LO Ttl LO T^ Ttl '^ lO ^^ TjH TJ1 TJi '^t' o fe.

'^ tí o o í3 I O O 00 00 (M TH O CO 00C0t^iO(N00

>^ l> 00 TíH Tti CO CO CM a: '^ r-^ t^ 05 CO CM Oi Ci CO (N ^ CO CO CO CO CM CO CM CM Tí^ CO CO CO '^ a i-H T—1 T—1 1—1 rH T—1 T—1 a ^1 ce I .PH^ i-i I o5 LO Tt< rH CM '—I Ttl i> CO CO OOCOCMCOOOrHLOOCO CM ^ ^ CM CM CM CM CM CM O^Tf^CMCM|>COrí^CO o.s

03 •+0 O) 00 CM oi TÍH O 00 CO "^ a> CM 1> 1> O O I 00 TH i bO tí CO CM TfH CO Tf^ CM CM CM CM r-i T-i Tin CO CO I CO TíH 03 03 ^ ^ o3 > <1 m <1 << o; 5 tí

pq tí 3 p

O) o; o3 rtí -13 tí C bD o3 tí o3 o o CO O O tí ^ PH lù bß 2 2 ^ o g ^ .„ o; O) ^ ^ ^ o "^ > 03 I' g O M ,ií I -t-^ í^ _r^ tí 03 .-H bC ^

TABLE 12.—Average days from seeding to first heading of 6 rice varieties representing 4 maturity groups, when sown in each of the 4 months of the planting period in the years from 1953 to 1961

Average days from seeding to first heading for plants Maturity group and variety sown in— Average March April May June

Number Early: Nato 103 88 80 76 87 Midseason: Sunbonnet 118 103 98 93 103 Toro 118 102 96 90 102 Medium-late : Blue Rose 137 115 102 86 110 R-D 136 121 115 105 119 Late: Rexoro 153 137 127 114 132

Average 128 111 103 94 _

seeding to maturity in April as compared with later than Sunbonnet when seeded early but be- March seeding was caused by greater temperature came earlier from subsequent seeding. differences that occur between April and later Early-maturing varieties do not show the char- seeding dates. The further reduction obtained acteristics of photoperiodic response, otherwise from May and June seedings probably was largely they would not head during the long days of June the result of photoperiodic response. Because of and July. Or possibly the critical length of the slow ripening late in the growing season, very late dark period is much shorter for them than for the seedings tend to lengthen the life cycle rather than more sensitive later varieties. reduce it. All varieties have a vegetative stage of develop- Seven varieties that mature in from 100 to 165 ment during which they are not photosensitive. days were sown at four successive dates each year The less sensitive midseason and later varieties from 1953 to 1961. The response of these varieties may have a longer vegetative stage than do the is shown in figure 25. The upper part of the curve early varieties. for each variety, representing the earlier part of Sensitive and short-season (early) varieties the planting period, tended to be steep, because may be seeded later than less sensitive midseason when the weather warmed in the spring, the length or late varieties. However, Arkrose and Caloro of the period from seeding to heading was rapidly are the only sensitive varieties now in production. reduced. Further reduction in time until heading Any of the less sensitive varieties can be used for for five of the seven varieties proceeded at a slower successive seedings because the plants mature in but comparatively uniform rate through the re- the order seeded, and the fields can be harvested mainder of the season. Consequently, the curves without conflict. indicating the response of those five varieties are Data obtained from date-of-seeding experi- roughly parallel. ments may be used to predict time of maturity Blue Eose and C.I. 6001 had steeper curves than of rice varieties. This information makes it pos- did the other five varieties. This reflected greater sible to schedule seedings so that different fields sensitivity to photoperiod. C.I. 6001 was later than and varieties may be harvested consecutively. E-D and Eexoro when seeded early but became The approximate dates of maturity of eight va- earlier from subsequent seeding. Blue Eose was rieties when sown at 10-day intervals, March 1 RICE IN THE UNITED STATEiS 63 to June 30, in southwest Louisiana are shown in early March to the end of June with expectation of profitable yields. In some years blast may be table 13. YIELDS IN EELATION TO SEEDING DATE.—Aver- serious during the warm humid weather in the age yields of variety groups by monthly seeding summer months. During years when this disease dates are summarized in table 14. Production is prevalent, there may be a reduction in stand of the early varieties varied but little because of of late-sown susceptible varieties. There may at date of seeding. Midseason and medium-late the same time be a reduced yield from early-sown, varieties were less productive from June seedings short-season varieties caused by the rotten-neck than from earlier seedings. Yields from May phase of blast. seedings of Eexoro and other late varieties were PLANT HEIGHT IN EELATION TO SEEDING DATE.— reduced. Yields from June seedings were un- Average plant heights of four groups of varie- profitably low or were complete failures. ties by monthly seeding ranges are summarized Probably Eexoro should not be seeded after in table 15. Midseason and late varieties were about May 20 at the latest {1^6), and April seed- taller than were early varieties. Midseason va- ing is preferred. Currently, no medium-late va- rieties seeded in June and late varieties seeded rieties, such as Blue Eose, are in production. in May and June were shorter than when seeded These have been replaced by early-maturing, earlier. Adair (^) reported that in Arkansas medium-grain varieties. The midseason varieties later seedings produced shorter straw, and Beach- Bluebonnet 50 and Sunbonnet may be seeded in ell {19) found a marked reduction in plant Louisiana as late as the first week of June with height of varieties that headed in response to a satisfactory results. Nato and other early varie- 10-hour photoperiod. ties may be seeded in Louisiana at any time from The reductions in average heights from later

3-20

6-10 6-30 7-20 8-9 8-29 9-18 10-8 DATE OF HEADING

FIGURE 25.—Heading date in relation to seeding date of 7 adapted varieties grown at Crowley, La., 1953-61. 64 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

seedings, as shown in table 15, correspond to re- not mean that short-strawed varieties necessarily duced yields in the respective months, as shown tend to be low yielding. However, among fields in table 14. of the same variety, plants with taller straw are A correlation of 0.25 between height and yield likely to be higher yielding than those with within individual varieties from seedings in the shorter straw. Exceptions to this generalization earlier months indicates that higher yields tend may occur when nitrogen fertilizer is applied at to be associated with greater plant heights from different times. early, medium, or late seeding dates. This does MILLING QUALITY IN KELATION TO SEEDING

TABLE 13.- -Approximate dates oj maturity of 8 rice varieties when sown at 10-day intervals in southwest Louisiana

Date of maturity for-

Seeding date Bluebonnet, Belle Patna Gulfrose Nato Sunbonnet, Texas Patna Rexoro or Toro

Mar. 1 July 13 "July <^^J 25 J»/L*XJuly 28 Aug. 13 Sept. 8___. Sept. 18, 10 17 27 3030__ 15 11. 20. 20 22___ Aug. 2 Aug. 4 22_ 15. 25. 30 26__. 4 9 27_ 21_ Oct. 2. Apr. 10 Aug. 2_ 13 16 31_ 27. 6. 20 9_ 21 22 Sept. 5 Oct. 3_ 13. 30 14. 27 30 15 7_ 19. May 10 21_ 31 Sept., 5 21 13_ 25. 20 29_ Sept. 4 15 Oct. 1 18. 30. 30 Sept. 5_ 12 24 11 25. Nov. 4. June 10_ 13_. 23 30 18. Nov. 3. 15. 20_ 26.. Oct. 4 Oct. 9 25_ 12_ 0). 30. Oct. 8__ 16 17 Nov. 4. 0) 0).

Failed to mature.

TABLE U.—Average yields oJ 3 rice varieties in each of 4 maturity groups Jrom seedings made in each of the 4 months of the planting period in the years from 1953 to 1961

Yields for maturity group Time of seeding Average Early Midseason Medium-late Late

Pounds per acre March 2 771 2,740 2, 634 2,521 2, 667 1 ' 2,674 2,744 2, 852 2,324 2,649 2,604 2,632 2,826 2, 172 2,568 2 62.*^ 2,395 2,452 1 1, 197 2 2, 167

Average 2,668 2,628 2, 691 3 2, 054 _.

1 Late varieties failed in 1954 and 1957. 2 Average of June seedings, late group excluded, 2,490. 3 Average of late varieties, June seeding excluded, 2,339. RICE IN THE UNITED STATEiS 65

DATE.—^Average percentages of head rice from in milling quality from June seedlings, probably varieties representative of four groups are sum- because of effects of low temperatures late in the marized according to monthly seeding date in season. table 16. No comparison can be made of the The early and medium-late, medium-grain vari- average milling quality of these groups because eties tended to give increasingly higher percent- of the lack of correspondence in grain type of ages of head rice from each successive seeding. the varieties within groups. Thus, apparently improved milling quality is a March seeding resulted in low percentages of major advantage of late seeding. head rice, probably because of higher temperatures during the maturation period. Eipening is has- Production of Seed Rice tened in hot weather, and more chalky grains are Origin of High-Quality Seed Rice produced. Also, rapid changes in temperature and moisture tend to cause checking of the grain. Com- According to Wise {108), new and superior vari- pared with March seedings, April and May seed- eties of crops make their intended contributions to ings gave improved milling quality. The average agriculture only "when the seed stocks of such percentage of head rice was 3 percent higher from crops reach the farmer varietally pure, in a viable May than from April seedings. condition, free of noxious weeds, in adequate quan- The average percentages of head rice from May tities and at a reasonable price." and June seedings were equal. However, the mid- To produce this high-quality seed, a grower must season and especially the late varieties were lower have a source of superior seed of a well-adapted

TABLE 15.—Average 'plant heights of 8 rice varieties in each oj I^. maturity groups^ from seedings made in each of the 4 months of the planting period in the years from 1953 to 1961

Plant heights for maturity group Time of seeding Average Early Midseason Medium-late Late

Inches

March 43 47 49 49 47 April- 44 47 50 49 48 May__ 44 47 49 48 47 June_- 45 46 45 45 45

Average 44 47 48 48 _.

TABLE 16.—Average percentages of head rice from samples of 8 rice varieties in each of 4 maturity groups^ from seedings made in each of the 4 months of the planting period in the years from 1958 to 1961

Head rice for maturity group Time of seeding Average Early Midseason Medium-late Late

March» _. 57 53 49 47 52 April 60 56 54 49 55 May 60 61 57 52 58 June 65 59 58 48 58 66 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE variety. Promising new strains are compared with tified seed, it is necessary to have clean land and standard varieties continually at the rice experi- to prevent mixtures in seeding, harvesting, and ment stations to determine their adaptation. When processing. Although careful roguing of all fields an experimental strain proves to be superior to to remove undesirable weeds, other crops, or off- standard varieties, it is increased for release to type plants increases the production costs, it is growers. necessary. Formerly each farmer could obtain a small The production of breeder and foundation seed amount of seed of a new or standard variety and is an integral part, of the cooperative rice-breed- thereafter produce his own seed. However, modern ing projects of the U.S. Department of Agricul- harvesting and processing methods, including bulk ture and the Agricultural Experiment Stations. drying and storage, have increased the possibility The production and certification of registered and of mixing. These methods and the use of more certified seed are not a part of the breeding pro- specific types of varieties that differ in maturity, gram. grain type, and processing and cooking quality BREEDER SEED PRODUCTION.—After an experi- have emphasized the need for sources of pure seed. mental variety of rice has been developed in the As a result, the seed certification program now in coordinated breeding program and has been effect in each major rice-producing State is an im- proved sufficiently outstanding, procedures are portant part of the rice industry. begun to purify it and to provide a seed supply In general, before a variety is approved by an for possible release to growers for commercial experiment station, it must be tested thoroughly production. (usually for a minimum of 3 years), and it must Procedures differ at the various experiment sta- show merit as a new variety in production, disease tions but the steps included usually are somewhat resistance, or some other outstanding character. as follows: (1) From 100 to 500 panicles are selected from the interior rows of nursery or Classes of Seed in a Certification Program field plots of the experimental variety; (2) these The classes of seed usually included in a certifi- panicles are weighed and the weight is added to cation program are breeder, foundation, registered, the plot weight so as not to cause inaccurate plot and certified. These are described by the Associa- yield reports; (3) each panicle is inspected and tion of Official Seed Certifying Agencies (3) as any having offtype seeds is discarded; (4) each follows : panicle typical of the variety is threshed indi- (1) Breeder seed is seed directly controlled by vidually; and (5) the seeds are placed in a sm^ill the originating or sponsoring plant breeding in- envelope. stitution, firm, or individual, and is the source for The next year the seed from each panicle the production of seed of the certified classes. that passed the screening test is sown in a single (2) Foundation seed shall be the progeny of row from 4 to 20 feet long and 12 to 24 inches Breeder, Select, or Foundation seed handled to apart. In some cases there are 3-foot alleys be- maintain specific genetic purity and identity. Pro- tween ranges of rows to facilitate careful inspec- duction must be acceptable to the certifying tion and roguing. The block of panicle rows of agency. each variety is isolated from those of varieties similar in maturity so as to eliminate natural (3) Registered seed shall be the progeny of crossing and subsequent segregation for offtypes. Breeder, Select, or Foundation seed handled under After the seedlings emerge, they are carefully procedures acceptable to the certifying agency to inspected at intervals to identify any offtype maintain satisfactory genetic purity and identity. plants or rows. Rows that show offtype plants or (4) Certified seed shall be the progeny of Breed- any apparent differences at any stage are removed er, Select, Foundation, or Eegistered seed so han- immediately or tagged for removal before harvest. dled as to maintain satisfactory genetic purity and In the early seedling stage, special emphasis is identity, and which has been acceptable to the given to noting albino seedlings or variable plant certifying agency. types. Later in the season special attention is paid For the production of the various classes of cer- to uniformity of vegetative growth and heading RICE IN THE UNITED STATES 67

within and among rows. During the ripening tion lower than foundation seed. Purification of period, careful observations are made to detect the variety could be continued and foundation seed differences in plant type, plant height, panicle could be released as soon as it becomes available. type, pubescence, color of apex or apiculus, and It may be desirable to check the processing grain type. If the entire group of panicle rows and cooking quality of the bulk seed from each appears uniform, the rows are harvested in bulk. family row used in the purification increase. Use- However, if considerable variability attributed ful quality tests are the alkali digestion (79) to genetic segregation is evident or if numerous and amylose tests (^i). The final bulk represent- offtype plants are found, then it may be neces- ing all the family lines should be grown in variety sary to make further selection of rows for puri- trials to determine the overall performance of the fication. This may be done by selecting from mass-selected strain. within a block of rows individual rows that ap- One method used to produce breeder seed of pear to be uniform in appearance and that typify established varieties is to select panicles from the the variety being increased. best available source of seed of that variety, such The grain from 30 to 50 or more such rows as a field being grown for production of foun- (families) may be harvested separately after 25 dation seed. Depending on the facilities available to 100 panicles are selected from each row. A and the amount of seed desired, individual number is assigned to each family for maintain- panicles may be selected in quantities varying ing its identity. The following year, from 10 to from 500 to possible 5,000. These panicles are care- 20 or more rows may be sown from the bulk seed fully inspected and those that are typical for the of each family, or a similar number of panicle variety are threshed individually and head rows rows from panicles saved from each family row are grown (fig. 26). may be used instead of the bulk. One procedure that has helped to eliminate An alternate method would be to select a few natural crosses with other varieties has been to panicles from all rows that appear to be typical seed the block of breeder panicle rows within the of the variety and grow three to six panicle rows area of a foundation field of the same variety. of each row. Each row must be examined carefully through- The panicles saved from each family row are out the growing season and atypical rows elimi- examined individually for offtypes and are grown nated. Failure to eliminate a row with a few off- by seeding each family in a group or block of three type plants will adulterate the seed produced, and to 25 or more panicle rows. As before, the rows it must be discarded. Typical rows are bulked. are observed for offtype or undesirable type plants The family method described for selecting new in the seedling and later stages. If several off- varieties may be necessary if a commercial variety types are found within a family or if the rows becomes badly mixed. But this method should be within a family tend to be variable or are not used as a last resort because severe mass selection typical of the variety, the entire family may be may result in genetic alteration of the original eliminated. If there still appears to be too much variety. variation in the material, it may be necessary to An alternate method used for the production again select individual rows from families most of breeder seed of established varities is to start similar in plant type. Several rows may be selected with the best source available and, depending on from each of several families, again identifying the amount of breeder seed desired, to carefully the families and subfamilies, and the subfamilies handpick a given quantity of seed to eliminate grown in panicle-row blocks the following year. grains that appear to be offtype or to have other Usually at this stage the material is suiSciently undesirable characteristics. This handpicked uniform, so that there are only a few offtypes to seed can be drilled thinly, possibly at one-fourth discard or eliminate. the normal rate, in rows 50 to 150 feet long, If a variety needs to be released as soon as spaced 12 or more inches apart. This increase possible because of a disease emergency, for ex- block then is observed very carefully at intervals ample, one procedure would be to increase and throughout the growing season, and offtype plants release seed at an earlier stage and at a designa- are eliminated. This method is much less time 68 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

FiouKE 26.—Breeder seed head-row block. consuming than the panicle-row method and ap- to produce the seed for storage is saved and stored pears to be quite satisfactory for well-established under conditions suitable for retaining the viability varieties that have been purified several times for more than 25 years. Each 5 to 10 years, or as previously. This method may be preferred to the needed, a part of this remnant seed may be sown to panicle-row method because with it there is less produce another supply of breeder seed identical chance of genetic alteration. to that originally stored. A system that was inaugurated at Beaumont, FOUNDATION SEED PRODUCTION.—Foundation Tex., in 1962 may eliminate growing breeder seed usually is the first-year increase from breeder seed panicle rows of a variety after a pure source seed. Foundation seed is produced on fields that of breeder seed is established. Under this sys- have not grown another variety or a lower class tem the jilant breeder produces a fairly large of the same variety during the 2 previous years. quantity of seed from panicle rows or family Preventing mixtures throughout the various blocks that is true to type for the variety. This phases of seed production requires very close seed is cleaned and put into 50- to 100-pound attention when several varieties are handled with containers and is placed in storage under condi- the same equipment. To facilitate roguing of tions suitable for maintaining the viability for foundation seed fields, a space is left every few at least 10 years. One or more units from stor- feet by stopping up one or more of the holes in age can be sown each succeeding year to produce the farm-type grain drill used for seeding. Such foundation seed. fields should be rogued several times during the The advantage of this method is that once a last part of the growing season. Insofar as pos- variety is purified, a continuing source of seed sible, foundation seed fields are managed to pro- of known purity is available. To carry this sys- duce satisfactory grain yields without excessive tem one additional step, a part of the seed used vegetative growth and to minimize lodging. It RICE IN THE UNITED STATElS 69 is impossible to satisfactorily rogue a field in foreign material ; (2) a screen with large perfora- which an appreciable amount of lodging has tions that removes any remaining sticks, stems, occurred. mud lumps, or large weed seeds; and (3) a finely The release of foundation seed to growers perforated screen that removes the finer broken usually is handled through a committee or seed rice grains, small weed seeds, and other small council or similar organization that allots the particles of foreign material. It may be necessary seed to carefully selected growers, often on the to put the rice through the cleaner a second time basis of rice acreage within a county or parish. to remove less easily separated material such as Sometimes the seed is turned over to a seed grow- shelled rice grains and weed seeds or an excess of ers organization or crop improvement associa- light-weight, underdeveloped grains. tion. Sometimes foundation seed is distributed The second step is a grain-length separation that directly to growers from the State agricultural may be accomplished either by a disk or an in- experiment station. For new varieties or for old dented cylinder-type machine. Small pockets or varieties in short supply, requested amounts of indents in revolving disks or cylinders retain the seed may be reduced in proportion to the amount shorter length grains (including broken grains that is requested. and hulled grains) slightly longer while the grain is lifted by the revolving disk or cylinder. A spe- Cleaning, Grading, and Processing Seed Rice cial compartment collects and eliminates the re- jected material from the grain. Cleaning and processing seed rice is an exacting The third step in the cleaning and grading proc- operation that requires specialized equipment. ess is a diameter or width separation that removes Where conditions and facilities permit, it is de- any large-diameter rice grains, weed seeds, or sirable to delay harvesting seed rice until the foreign materials. For this operation either a ver- moisture content is below 20 percent. If the com- tical screen or a perforated cylinder grader is used. bine harvester is carefully adjusted, the rough rice The grains of normal diameter pass through the coming from "clean" fields may be relatively free screen perforations while the grains of larger di- of stems, weed seeds, and trash so that it can be ameter, the weed seeds, or forign materials are unloaded and safely elevated directly into aerated retained and thus removed from the sample. For bins without prior aspirating or scalping. In such successful length and width separations, the sam- case the seed rice is placed in hopper-bottom bins ple should be free from sticks, stems, and other and aerated with an excess of air to dry the grain foreign materials because such materials will inter- or at least keep it from heating. Where necessary, fere with the proper functioning of the machine. the air may be heated to facilitate drying of the The length and diameter grading of seed rice seed rice to a moisture level sufficiently low for safe has been extremely useful in removing the larger storage. If rice coming from the field contains con- diameter red rice grains from seed of long-grain siderable foreign material, it may be advisable to varieties. The use of these graders has been im- partly clean it w4th a scalper-aspirator machine portant in the control of red rice. In medium- and before putting it in the bin for aeration. In some short-grain varieties, the only means of red rice locations, facilities and conditions require that the control is the use of red rice free seed and land rough rice be dried before it can be safely stored. because no economical method of separation has This drying may require that the rice be passed as yet been devised. In some instances long-grain through the drier several times. Frequently the red rice types have occurred in the long-grain rice is aspirated between passes to remove foreign varieties. Whenever a lot of long-grain seed rice matter and light-weight, immature grains. Ex- has a mixture of long-grain red rice, it should be treme care must be exercised to prevent mechanical discarded immediately and not used for seed. The mixing if the drying and cleaning facilities are propagation of seed containing long-grain red rice used for more than one variety of rice. will soon result in a wide infestation of the soil The first step in the cleaning and grading proc- ess is to put the rice through a fanning mill, which with long-grain red rice strains and will has the following parts : (1) a wind aspirator that further complicate the maintenance of pure seed removes light grain, hulls, and other light-weight production. 70 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

Following the diameter-grading operation, the tion, seven 35-hundredweight bins are available seed rice usually is treated with a fungicide and for small lots of breeder seed. All bins are of steel often with an insecticide and placed in well- construction and are equipped with aeration. marked 100-pound bags. The seed is then ready Small portable cleaning units without elevators for distribution to selected growers. or conveyors frequently are used for cleaning small Each of the rice experiment stations has a plant lots of breeder seed. These small lots usually are for processing seed lice. These plants are designed dried on sack driers or small laboratory driers to to make the installations as nearly self-cleaning as further reduce hazards of mixing. Various sizes possible. They are being improved constantly as and types of seed-cleaning and handling equip- new and improved equipment and methods are ment are described in detail by Harmond, Klein, developed. Gravity movement of bulk seed is used and Brandenburg (4£). whenever possible. Steel or metal-lined bins with smooth walls and gravity flow metal hopper bot- Standards for Seed Certification toms simplify cleaning the facilities. Belt convey- The standards for field inspection and labora- ors are preferred to other types because of ease of tory analysis of seed samples for seed certification cleaning. are summarized in tables 17 and 18. These are the The seed rice processing plant at Stuttgart, Ark., standards set forth by the Association of Official was described by Williams (106). It includes a Seed Certifying Agencies (3). All rice-producing concrete dump pit, a truck hoist, a 60-hundred- States use these standards as the minimum re- weight-per-hour drier, two bucket-type elevators, quirements for seed rice. ten 450-hundredweight bins, two half-size working Each State certifying agency designates the bins, cleaning equipment, a seed treater, and 3,500 range of types acceptable as "other varieties" and square feet of sack storage space (fig. 27). In addi- the names of objectionable and noxious Aveeds. Eed

FIGURE 27.—Seed rice processing plant at Stuttgart, Ark. RICE IN THE UNITED STATEiS 71

TABLE 17.—Standards jor field inspection oj rice

standards for each class Factor Foundation Registered Certified

Other varieties 1:10,000 plants 1:5,000 plants 1:1,000 plants. Red Rice None None 1:100,000 plants. Objectionable or noxious weeds None None -•-_ None. Disease affecting quality of seed or transmissible None None None. through planting stock that can be controlled with seed treatment.

TABLE 18.—Standards jor cleaned seed oj rice

Standards for each class Factor Foundation Registered Certified

Pure seed (minimum) 98.0 percent 98.0 percent 98.0 percent. Inert matter (maximum)__ 2.0 percent 2.0 percent 2.0 percent. Weed seeds (maximum) 0.05 percent 0.05 percent 0.1 percent. Objectionable or noxious weed seeds None None None. Total other crop seeds (maximum) : Other varieties (maximum) 0.05 percent 0.10 percent 0.21 percent. Other kinds (maximum) None None 0.01 percent. Germination (minimum) 80.0 percent 80.0 percent 80.0 percent. Moisture (maximum) 14.0 percent 14.0 percent 14.0 percent.

rice is not allowed in Foundation or Registered Selected References Seed and only one plant per 100,000 plants in the field or one seed per pound in Certified seed. (1) ANONYMOUS. 1960. GULFROSE ... A NEW RICE VARIETY RESIST- In order for the rice to be eligible for certifica- ANT TO HOJA BLANCA DISEASE. TeX. Agr. tion, the seed rice field must not have had another Expt. Sta. Leaflet 484, 4 pp. rice variety or the same variety of lower class (2) growing on it the previous year. The field also 1961. BELLE PATNA ... A NEW SHORT-SEASON, must be isolated from other ricefields. The field LONG-GRAIN RICE VARIETY. TeX. Agr. Expt. Sta. Leaflet L-512, 4 pp. must be clearly separated from other fields by a (3) ditch, levee, roadway, fence, or barren strip at 1971. CERTIFICATION HANDBOOK. Assoc. Off. Seed least 10 feet wide if the adjoining field is the same Certif. Agencies. 168 -f 6 pp. [Mimeo- variety and class of seed. If of another variety or graphed.] (4) AD AIR, C. R. the same variety of a lower class of seed, at least 1940. EFFECT OF TIME OF SEEDING ON YIELD, MILL- 100 feet shall separate the two. ING QUALITY, AND OTHER CHARACTERS IN RICE. Specific requirements and standards are estab- Amer. Soc. Agron. Jour. 32: 697-706. (5) lished for each State and a list of these is avail- 1941. INHERITANCE IN RICE OF REACTION TO HEL- able from each of the official certifying agencies MINTHOSPORIUM ORYZAE AND CERCOSPORA {S). In general, they are concerned with applica- ORYZAE. U.S. Dept. Agr. Tech. Bui. 772. tion procedures, field and harvest inspection, post- 19 pp. (6) harvest seed movement, seed processing, and offi- 1968. TESTING RICE SEEDLINGS FOR COLD WATER TOL- cial sampling. ERANCE. Crop Sei. 8: 264-265. 72 AGRICULTURE HAISTDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

(7) - and CRALLEY, E. M. (20) 1950. 1949 RICE YIELD AND DISEASE CONTROL TESTS. 1959. RICE. In Matz, S. A., ed., The Chemistry Ark. Agr. Expt. Sta. Rpt. Ser. 15, 20 pp. and Technology of Cereals as Food and (8) and JONES, J. W. Feed, pp. 137-176. Avi Pub. Co., Inc., 1946. EFFECT OF ENVIRONMENT ON THE CHARACTER- Westport, Conn. ISTICS OF PLANTS SURVIVING IN BULK HYBRID (21) and HALICK, J. V. POPULATIONS OF RICE. Amer. Soc. Agron. 1957. BREEDING FOR IMPROVED MILLING, PROCESSING Jour. 38: 708-716. AND COOKING CHARACTERISTICS OF RICE. In- (9) MILLER, M. D., and BEACHELL, H. M. tematl. Rice Comn. Newsletter 6(2) : 1-7. 1962. RICE IMPROVEMENT AND CULTURE IN THE (22) and HALICK, J. V. UNITED STATES. Adv. in Agron. 14: 61-108. 1957. PROCESSING AND COOKING QUALITIES OF RICE (10) ANDERSON, A. L., HENRY, B. W., and TULLíS, E. C. AND METHODS FOR THEIR DETERMINATION. 1947. FACTORS AFFECTING INFECTIVITY, SPREAD, AND Presented Sixth meeting. Working Party PERSISTENCE OF PIRICULARIA ORYZAE CAV. on Rice Breeding, IRC, FAO, United Na- Phytopathology 37: 94-110. tions. Vercelli, Italy, 9 pp. (11) ATEN, A., and FAUNCE, A. D. (23) and JENNINGS, P. R. 1953. EQUIPMENT FOR THE PROCESSING OF RICE. 1961. MODE OF INHERITANCE OF HOJA BLANCA RE- FAO Devlpmt. Paper 27, 55 pp. United SISTANCE IN RICE. Rice Tech. Working Nations. Group Proc, pp. 11-12.

(12) ATKINS, J. G., and ADAIR, C. R. (24) JoDON, N. E., JOHNSTON, T. H., and others. 1957. RECENT DISCOVERY OF HOJA BLANCA, A NEW 1959. BREEDING RICE VARIETIES FOR RESISTANCE TO RICE DISEASE IN FLORIDA, AND VARIETAL RE- THE VIRUS DISEASE HOJA BLANCA. Inter- SISTANCE TESTS IN CUBA AND VENEZUELA. nat!. Rice Comn. Newsletter 8(3) : 6-9. U.S. Dept. Agr. Plant Dis. Rptr. 41: 911- (25) SCOTT, J. E., EVATT, N. S., and others. 915. 1961. BELLE PATNA. Rlce Jour. 64(6) : 6-8, 24- (13) BEACHELL, H. M., and CRANE, L. E. 26.

1956. REACTION OF RICE VARIETIES TO STRAIGHT- (26) BoLLicH, C. N,, ATKINS, J. G., SCOTT, J. E., and HEAD. Tex. Agr. Expt. Sta. Prog. Rpt. WEBB, B. D. 1865, 2 pp. 1966. DAWN ... A BLAST RESISTANT, EARLY MATUR- (14) BEACHELL, H. M., and CRANE, L. E. ING, LONG GRAIN RICE VARIETY. RicC JOUr. 1957. TESTING AND BREEDING AMERICAN RICE VARI- 69(4) : 14, 16, 18, 20. ETIES FOR RESISTANCE TO STRAIGHTHEAD. (27) ATKINS, J. G., SCOTT, J. E., and WEBB, B. D. Internatl. Rice Comn. Newsletter 6(2) : 1968. REGISTRATION OF DAWN RICE. (Rcgistr. No. 12-15. 33.) Crop Sei. 8: 400. (15) BOLLICH, C. N., JOHNSTON, T. H., and others. (28) SCOTT, J. E., WEBB, B. D., and ATKINS, J. G. 1963. BREEDING FOR BLAST RESISTANCE IN THE 1966. BLUEBELLE ... A LODGING RESISTANT, VERY UNITED STATES. Blast Symp. Proc, Inter- EARLY MATURING, LONG GRAIN RICE VARIETY natl. Rice Res. Inst., Los Banos, Philip- RELEASED IN TEXAS. RicC JOUr. 69(1) I 13- pines. July. Pp. 333-341. 17. (16) . and JODON, N. E. (29) SCOTT, J. E., WEBB, B. D., and ATKINS, J. G. 1963. ASPECTS OF BREEDING RICE FOR RESISTANCE 1968. REGISTRATION OF BLUEBELLE RICE. (Registr. TO DISEASES, PARTICULARLY BLAST (PIRICU- No. 32.) Crop Sei. 8: 400-401. LARIA ORYZAE). Internatl. Rice Comn. (30) BoRASio, L., and GARIBOLDI, F. Newsletter, Spec, issue, 10th Pacific Sei. 1957. ILLUSTRATED GLOSSARY OF RICE PROCESSING Cong. Symp., pp. 41-52. EQUIPMENT. Food and Agr. Organ, pp. 65, (17) , and TODD, E. H. United Nations. 1959. WHITE TIP DISEASE OF RICE. Ill YIELD (31) BOWMAN, D. H. TESTES AND VARIETAL RESISTANCE. PhytO- 1966. DAWN . . . ITS PERFORMANCE IN MISSISSIPPI. pathology 49; 189-191. Rice Jour. 69(4) : 23. (18) AuTREY, H. S., GRIGORIEFF, W. W., ALTSCHUL, A. M. (32) CIIAMBLISS, C. E., and JENKINS, J. M. and HoGAN, J. T. 1923. SOME NEW VARIETIES OF RICE. U.S. Dept. 1955. RICE MILLING EFFECTS OF MILLING CONDI- Agr. Dept. Bui. 1127,17 pp. TIONS ON BREAKAGE OF RICE GRAINS. Jour. (33) CHILTON, S. J. P., and TULLíS, E. C. Agr. and Food Chem. 3: 593-599. 1^>46. A NEW RACE OF CERCOSPORA ORYZAE IN RICE. (19) BEACHELL, H. M. Phytopathology 36: 950-952. 1943. EFFECT OF PHOTOPERIOD ON RICE VARIETIES (34) CRALLEY, E. M. GROWN IN THE FIELD. Jour. Agr. Res. 66: 1949. WHITE TIP OF RICE. (Abstract) Phyto- 325-340. pathology 39: 5. RICE IN THE UNITED STATElS 73

(35) CRANE, L. E. (50) 1959. NEW RICE VARIETY, "GULFROSE," INTRODUCED. 1961. PERFORMANCE OF RICE VARIETIES IN 1961. Rice Jour. 62(13) : 2-3, 17. 53rd Ann. Prog. Rpt., La. Rice Expt. Sta., 162 pp. [Processed.] (36) DAVIS, L. L. 1950. CALIFORNIA RICE PRODUCTION. CaÜf. Agr. (51) Ext. Ser. Cir. 163, 55 pp. 1965. SATURN RICE. (Registr. No. 29.) Crop Sei.

(37) FRAPS, G. S. 5: 288. 1916. THE COMPOSITION OF RICE AND ITS BY-PROD- (52) and CHILTON, S. J. P. UCTS. Tex. Agr. Expt. Sta. Bui. 191, 41 pp- 1946. SOME CHARACTERS INHERITED INDEPEND-

(38) GEDDES, W. F. ENTLY OF REACTION TO PHYSIOLOGIC RACES 1951. RICE MILLING. Ifi Jacobs, M. B., ed., The OF CERCESPORA ORYZA IN RICE. Amcr. SOC. Chemistry and Technology of Food and Agron. Jour. 38: 864-872. Feed Products. 3 v. Interscience Pub., Inc., (53) and DE LA HoussAYE, D. A. New York. 1949. RICE VARIETIES FOR LOUISIANA. La. Agr. (39) HALICK, J. V., BEACHELL, H. M., STANSEL, J. W., Expt. sta. Bui. 436,15 pp. and KRAMER, H. H. (54) and MCILRATH, W. O. 1960. A NOTE ON THE DETERMINATION OF GELATINI- 1971. RESPONSE OF RICE TO TIME OF SEEDING IN ZATION TEMPERATURES OF RICE VARIETIES. LOUISIANA. La. Agr. Expt. Sta. Bui. 649, Cereal Chem. 37: 670^672. 27 pp. (40) and KELLY, V. J. (55) RYKER, T. C, and CHILTON, S. J. P. 1959. GELATINIZATION AND PASTING CHARACTERIS- 1944. INHERITANCE OF REACTION TO PHYSIOLOGIC TICS OF RICE VARIETIES AS RELATED TO COOK- ING BEHAVIOR. Cereal Chem. 36: 91-97. RACES or CERCOSPORA ORYZA IN RICE. Amer. Soc Agron. Jour. 36: 407-507. (41) and KENEASTER, K. K. 1956. THE USE OF A STARCH-IODINE-BLUE TEST AS A (56) and SoNNiER, E. A. QUALITY INDICATOR OF WHITE MILLED RICE. 1966. DAWN . . . PERFORMANCE AND SEED PRODUC- Cereal Chem. 33: 315-319. TION IN LOUISIANA. Rice Jour. 69(4) : 20. (42) HARMOND, J. E., KLEIN, L. M., and BRANDENBURG, 22. N.R. (57) JOHNSTON, T. H. 1961. SEED CLEANING AND HANDLING. U.S. Dept. 1958. REGISTRATION OF RICE VARIETIES. AgrOH. Agr., Agr. Handb. 179, 38 pp. Jour. 50: 694-700. (43) JENKINS, J. M. (58) ADAIR, C. R., TEMPLETON, G. E., and others. 1936. EFFECT OF DATE OF SEEDING ON THE LENGTH 1963. NOVA AND VEGOLD . . . NEW RICE VARIETIES. OF THE GROWING PERIOD OF RICE. La. Agr. Ark. Agr. Expt. Sta. Bui. 675, 23 pp. Expt. Sta. Bui. 277, 7 pp. (59) and CRALLEY, E. M. (44) and JONES, J. W. 1955. RICE VARIETIES AND THEIR YIELDS IN ARKAN- 1944. RESULTS OF EXPERIMENTS WITH RICE IN SAS, 1948-1954. Ark. Agr. Expt. Sta. Rpt. LOUISIANA. La. Agr. Expt. Sta. Bui. 384, Ser. 49, 20 pp. 39 pp. CRALLEY, E. M., and HENRY, S. E. (45) JODON, N. E. (60) 1959. PERFORMANCE OF RICE VARIETIES IN ARKAN- 1938. EXPERIMENTS ON ARTIFICIAL HYBRIDIZATION OF RICE. Amer. Soc. Agron. Jour. 30: 294- SAS, 1953-1958. Ark. Agr. Expt. Sta. Rpt. 305. Ser. 85, 31 pp. (46) (61) TEMPLETON, G. E., and ATKINS, J. G. 1953. GROWING PERIOD OF LEADING RICE VARIETIES 1968. REGISTRATION OF NOVA 66 RICE. (Registr. WHEN SOWN ON DIFFERENT DATES. La. Agr. No. 30.) Crop Sei. 8:399. Expt. sta. Bui. 476, 8 pp. (62) TEMPLETON, G. E., SIMMS, J. L., and others. (47) 1966. DAWN RICE . . . ITS PERFORMANCE IN AR- 1955. IMPROVING VARIETIES OF RICE AND SORGHUM. KANSAS. Rice Jour. 69(4) : 2&-28. 47th Ann. Prog. Rpt., La. Rice Expt. Sta., (63) TEMPLETON, G. E., SIMS, J. L., and others. pp. 83-87. 1966. PERFORMANCE IN ARKAIfSAS OF NOVA 66 AND (48) OTHER MEDIUM-GRAIN VARIETIES, 1960-1965. 1955. SUNBONNET AND TORO. 2 NEW MIDSEASON Ark. Agr. Expt. Sta. Rpt. Ser. 148, 24 pp. LONG-GRAIN RICE VARIETIES. La. Agr. Expt. (64) TEMPLETON, G. E., WEBB, B. D., and others. Sta. Bui. 499,12 pp. 1967. PERFORMANCE IN ARKANSAS OF STARBONNET (49) AND OTHER LONG-GRAIN RICE VARIETIES, 1962 1957. NATO: AN EARLY MEDIUM-GRAIN RICE. La. to 1966. Ark. Agr. Expt. Sta. Rpt. Ser. Agr. Expt. Sta. Cir. 47, 14 pp. 160, 26 pp. 74 AGRICULTURE HANDBOOK NO. 2 8 9, U.S. DEPT. OF AGRICULTUREi

(65) - TEMPLETON, G. E., WELLS, J. P., and HENRY, (79) LITTLE, RUBY R., HILDER, GRACE B., and DAWSON, S. E. ELSIE H. 1962. NORTHROSE, A NEW SPECLA.L-PURPOSE RICE 1958. DIFFERENTIAL EFFECT OF DILUTE ALKALI ON VARIETY FOR ARKANSAS. RiCG JoUr. 65(8) : 2 5 VARIETIES OF MILLED WHITE RICE. Cereal 10, 12-14, 18-19. Chem. 35: 111-126. (66) TEMPLETON, G. E., WELLS, J. P., and HENRY, (80) MASTENBROEK, J. J., and ADAIR, C. R. S. E. 1970. REGISTRATION OF CS-M3 RICE. (Registr. No. 1962. NORTHROSE RICE A SPECIAL-PURPOSE, STIFF- 34.) Crop Sei. 10 : 728. STRAWED EARLY VARIETY FOR ARKANSAS. (81) NAGAI, I. Ark. Farm Res. 11(2) : 2. 1959. JAPóNICA RICE, ITS BREEDING AND CULTURE. (67) WEBB, B. D., and EVANS, K. O. 843 pp. Yokendo Ltd., Tokyo. 1968. REGISTRATION OF STARBONNET RICE. (Reg- (82) NELSON, MARTIN, and ADAIR, C. R. istr. No. 31. ) Crop Sei. 8: 399. 1940. RICE VARIETY EXPERIMENTS IN ARKANSAS. (68) JONES, J. W. Ark. Agr. Expt. Sta. Bui. 403, 28 pp. 1923. RICE EXPERIMENTS AT THE BIGGS RICE FIELD (83) ORMROD, D. P., and BUNTER, W. A., JR. STATION IN CALIFORNIA. U.S. Dept. Agr. 1961. THE EVALUATION OF RICE VARIETIES FOR COLD Dept. Bui. 1155, 60 pp. WATER TOLERANCE. Agrou. Jour. 53: 133- (69) 134. 1936. IMPROVEMENT IN RICE. U.S. Dept. Agr. (84) POEHLMAN, J. M. Yearbook 1936: 415-454. 1959. BREEDING RICE. VARIETAL HISTORY OF RICE (70) IN THE UNITED STATES. His Breeding Field 1938. THE "ALKALI TEST" AS A QUALITY INDICATOR Crops, pp. 174-189. Henry Holt and Co., OF MILLED RICE. Amer. Soc. Agron. Jour. Inc., New York. 30: 960-967. (85) RAMIAH, K. (71) ADAIR, C. R., BEACHELL, H. M., and others. 1933. INHERITANCE OF FLOWERING DURATION IN 1953. RICE VARIETIES AND THEIR YIELDS IN THE RICE (ORYZA SATIVA L. ). Indian Jour. Agr. UNITED STATES 1939-50. U.S. Dept. Agr. Sei. 3: 377-410. Cir. 915, 29 pp. (86) RAO, B. B., VASUDEVA MURTHY, A. R., and SUBRAH- (72) JENKINS, J. M., NELSON, MARTIN, and others. MANYA, R. S. 1941. RICE VARIETIES AND THEIR COMPARATIVE 1952. THE AMYLOSE AND THE AMYLOPECTIN CON- YIELDS IN THE UNITED STATES. U.S. Dept. TENTS OF RICE AND THEIR INFLUENCE ON Agr. Cir. 612, 34 pp. THE COOKING QUALITY OF THE CEREAL. In- (73) JENKINS, J. M., WYCHE, R. H., and NELSON, dian Acad. Sei. Proc. 36B : 70--80. MARTIN. (87) REYNOLDS, E. B. 1938. RICE CULTURE IN THE SOUTHERN STATES. 1954. RESEARCH ON RICE PRODUCTION IN TEXAS. U.S. Dept. Agr. Farmers' Bui. 1808, 29 pp. Tex. Agr. Expt. Sta. Bui. 775, 29 pp. (74) KARON, M. L., and ADAMS, MABELLE E. (88) RICE MILLERS ASSOCIATION 1949. HYGROSCOPIC EQUILIBRIUM OF RICE AND RICE 1970. RICE ACREAGE IN THE UNITED STATES 1970. FRACTIONS. Cereal Chem. 26: 1-12. Rice Jour. 73(8) : 10-11. (75) KESTER, E. B. (89) RYKER, T. C. 1959. RICE PROCESSING. In Matz, S. A., ed., The 1943. PHYSIOLOGIC SPECIALIZATION IN CERCOSPORA Chemistry and Technology of Cereals as ORYZAE. Phytopathology 33 : 70-74. Food and Feed, pp. 427-461. Avi Pub. Co., (90) Inc., Westport, Oonn. 1947. (76) KiK, M. C, and WILLIAMS, R. R. NEW PATHOGENIC RACES OF CERCOSPORA ORY- ZAE AFFECTING RICE. (Abstract) Phyto- 1945. THE NUTRITIONAL IMPROVEMENT OF WHITE RICE. Nati. Res. Council Bui. 112, 76 pp. pathology 37: 19-20. (77) KING, B. M. (91) and CowART, L. E. 1937. THE UTILIZATION OF WABASH CLAY (GUMBO) 1948. DEVELOPMENT OF CERCOSPORA-RESISTANT SOILS IN CROP PRODUCTION. Mo. Agr. Expt. STRAINS OF RICE. (Abstract) Phytopathol- Sta. Res. Bui. 254, 42 pp. ogy 38: 23. (78) LATTERELL, FRANCES M., TULLíS, E. C, and COLLIER, (92) SCOTT, J. E., WEBB, B. D., and BEACHELL, H. M. J. W. 1964. RICE TEST-TUBE MILLER. Crop. Sei. 4: 231. 1960. PHYSIOLOGICAL RACES OF PIRICULARIA ORYZAE (93) WEBB, B. D., and BEACHELL, H. M. CAv. U.S. Dept. Agr. Plant Dis. Rptr. 44 : 1964. SMALL SAMPLE RICE POLISHING MACHINE. 67^-683. Crop. Sei. 4: 232. RICE IN THE UNITED STATES 75

(94) SIMS, J. L., HALL, V. L., JOHNSTON, T. H., and (102) WEBB, B. D., and ADAIR, C. R. BLACKMON, B. G. 1970. LABORATORY PARBOILING APPARATUS AND 1967. EFFECT OF RATES AND TIMING OF MIDSEASON METHODS OF EVALUATING PARBOIL-CANNING

NITROGEN APPLICATIONS ON PERFORMANCE OF STABILITY OF RICE. Cereal Chem. 47 : 708- SHORT-SEASON RICE VARIETIES, 1964-1965. 714. Ark. Agr. Expt. Sta. Rpt. Ser. 154, 24 pp. (103) WELLS, B. R., and JOHNSTON, T. H. (95) SMITH, W. D. 1970. DIFFERENTIAL RESPONSE OF RICE VARIETIES TO 1955. THE USE OF THE CARTER DOCKAGE TESTER TO TIMING OF MIDSEASON NITROGEN APPLICA- REMOVE' WEED SEEDS AND OTHER FOREIGN TIONS. Agron. Jour. 62: 608-612. MATERIAL FROM ROUGH RICE Rice JOUr. 58(9) : 26-27. (104) SIMS, J. L., JOHNSTON, T. H., and HALL, V. L. (96) 1970. RESPONSE OF SOME LONG-GRAIN RICE VARIETIES 1955. THE USE OF THE MCGILL SHELTER FOR RE- TO TIMING OF MIDSEASON NITROGEN APPLICA- MOVING HULLS FROM ROUGH RICE. Rice TIONS, 1964 to 1967. Ark. Agr. Expt. Sta. Jour. 58(10) : 20. Rpt. Ser. 184, 34 pp. (97) (105) WELLS, D. G., and CAFFEY, H. R. 1955. THE USE OF THE MCGILL MILLER FOR MILL- 1956. SCISSOR EMASCULATION OF WHEAT AND BAR- ING SAMPLES OF RICE. Rico Jour. 58(11) : LEY. Agron. Jour. 48: 496-499. 20. (106) WILLIAMS, F. J. (98) 1957. AT THE RICE BRANCH EXPERIMENT STATION 1955. THE DETERMINATION OF THE ESTIMATE OF ... A NEW SEED PROCESSING PLANT. Ark. HEAD RICE AND OF TOTAL YIELD WITH THE USE OF THE SIZING DEVICE. Rice JOUr. Farm Res. 6(5) : 2. 58(12) : 9. (107) WILLIAMS, V. R., Wu, W. T., TSAI, H. Y., and (99) TODD, E. H., and BEACHELL, H. M. BATES, H. G. 1954. STRAIGHTHEAD OF RICE AS INFLUENCED BY 1958. VARIETAL DIFFERENCES IN AMYLOSE CONTENT VARIETY AND IRRIGATION PRACTICES. TeX. OF RICE STARCH. Agr. Food Chem. 6(1) : Agr. Expt. sta. Prog. Rpt. 1650, 3 pp. 47-48. (100) WARTH, F. J., and DARABSETT, D. B. (108) WISE, L. N. 1914. THE FRACTIONAL UQUEFACTION OF RICE 1954. RESEARCH IN SEED PROCESSING. Amer. Soc. STARCH. Mem. Dept. Agr. India, Chem. Agron. 1954 Annual Meetings Agron. Abs., Ser. 3: 135-147. p. 91. (101) WAYNE, T. B. 1930. MODERN RICE MILLING AS PRACTICED IN THE (109) WITTE, G. C, Jr. WORLD'S LARGEST COMPLETE MILL. Food 1970. RICE MILLING IN THE UNITED STATES. Bul. Indus. 2: 492-495. Assoc. Operative Millers, pp. 3147-3159. SOILS AND FERTILIZERS

By D. S. MiKKELSEN and N. S. EVATT

Types of Soils Used for Rice Production tion, and the chemical properties of the soil all influence this pH change. Rice, a semiaquatic plant, benefits from ñooded Not only do pH changes occur when a rice soil is soil conditions during part or all of the growing flooded, but the continuous use of ammonium season. Flooding is essential for optimum grain sources of nitrogen such as ammonium sulfate, yields. Because of this water requirement, the ideal urea, and anhydrous and aqua ammonia also leaves soil types for rice production are those that con- an acid residue in the soil. The soil pH often de- serve water. Usually clay, clay loams, silty clay creases as much as 2 pH units where these nitrogen loams, or silt loams are considered most desirable sources have been used in rice production over a because they prevent excessive seepage losses. period of years. If the soil pH declines below pH 5, Other soils, including organic soils, can be used if applications of liming materials may be profitable. they have a claypan or hardpan that can maintain Soils high in pH due to their high lime content up to 6 inches of floodwater. Flooded soils, used or high sodium saturation often cause problems in in mechanized rice production, must also be able rice production by reducing the availability and to support the weight of equipment used to apply utilization of plant nutrients. Problems of zinc and fertilizers, herbicides, insecticides and grain com- iron deficiency are more frequent on these soils, bines and bankout wagons at harvesttime. Fields because of nutritional problems detrimental to should be laid out in such a manner that they are good stand establishment. Soils high in exchange- easy to irrigate. Surface drainage must be possible able sodium usually have a poor physical condi- in the event that midseason drainage or drainage tion due to dispersion of the soil colloids. These before harvesttime is required. soils often retain moisture longer than unaffected Soils that are well supplied with essential nutri- soils and cause problems when mechanical equip- ents are most desirable for the crop, although high ment is moved over the soil. levels of production are seldom possible without Salinity problems are sometimes encountered the application of nitrogen fertilizer. in areas where soluble salts have accumulated or Rice does not have an unusual soil pH require- where a poor quality of irrigation water is used. ment and grows well at the same range of reactions Sea water that is intruded into rivers and water- suited to most other field crops. A soil pH between ways or that is brought in by storms of hurricane 5 and 7.5 provides a good range for the growth of force may also be a source of excess salinity. Be- rice in all producing areas. In this range of pH cause rice can grow in flooded soils, it is occasion- values, nutrient availability is generally favorable ally used as a reclamation crop. Varieties differ and ions in harmful amounts such as sodium, in tolerance to salinity, but all are affected by the which may occur at high pH's, and aluminum, salt concentration of the water and the soil. The iron, and manganese, which may occur at low pH's, effects of salinity on rice depend somewhat on its are not likely to be a problem. The pH of the soil stage of development when it is exposed to saline after flooding, whether it be alkaline or acid in conditions. Results of field experiments have in- reaction, shifts temporarily toward neutrality dicated that high soil salinity occurring at plant- while the soil is flooded. Soils may shift from 0.5 ing time may seriously impair yield by reducing to 2.0 pH units, depending upon how strongly acid germination and stand establishment. Rice is most or alkaline they are, whereas soils initially neutral tolerant during germination and is most sensitive may not change appreciably. The soil organic mat- during the 1- to 2-leaf stage. Salt tolerance ap- ter content, the rate of organic matter decomjDOsi- parently increases during the tillering and elonga- 76 RICE IN THE UNITED STATEIS 77 tion stages but decreases during the flowering ditions this soil is poorly drained, and the natural stage. vegetation is hardwood and cypress. In the virgin Most rice soils, aften referred to as heavy soils state it is well supplied with nitrogen and phos- because of their high clay and silt content, pre- phorus, and it is slightly acid to neutral in re- sent special management problems. These in- action. In recent years, Sharkey clay has been clude tillage and seedbed preparation, maintenance widely used for rice in the Mississippi Valley. of organic matter and soil structure, adequate OTHER SOILS.—Other soils sometimes used for drainage for essential mechanized rice operations rice production in Arkansas are Waverly silt loam, and for other crops planted in rotation, green and Waverly, Miller, and Portland clay soils. Be- manure crops, fertilizer application, and weed con- cause these soils usually have poor natural drain- trol. Management of rice soils is discussed in the age, they have come into use for rice production section "Culture," p. 88. only in recent years. The soils most widely used in rice production in the United States are of alluvial origin (SS). Louisiana Brief descriptions of the soils in the principal rice- According to Walker and Miears {S6)j the growing States are given below. principal soils used for rice production in Loui- siana are the Crowley, Midland, and Beaumont Arkansas soils. CROWLEY SOILS.—Soils of this type found in Crowley silt loam, Calhoun silt loam, and Louisiana are the same as soils of this type in Sharkey clay are the principal soils used for rice Arkansas. production in Arkansas (^7), but other soils are MIDLAND SOILS.—The Midland soils, w^hich gen- sometimes used. erally are alluvium deposited by the Red and the CROWLEY SILT LOAM.—The surface of Crowley Mississippi Rivers, are deep and poorly drained silt loam is gray to brown. It is underlain with (5). The surface soil is gray to dark gray and a gray or yellowish-gray silt loam that changes strongly acid. The subsoil is strong-brown to light with depth into a gray silty clay and finally into olive-brown, heavy, silty clay mottled with gray. a heavy clay usually mottled with yellow and red. It is very strongly acid to mildly alkaline. Crowley silt loam in its virgin state has a fairly MIDLAND-CROWLEY MIXED SOILS.—Soils that are high organic matter content, but it is low in phos- a mixture of Midland silt loam. Midland silty clay phorus and is strongly acid. loam, and Crowley silt loam are used for rice in CALHOUN SILT LOAM.—Calhoun silt loam is a Louisiana. Feitility and organic matter content light-gray to almost white shallow soil under- of the soils are moderate, and surface runoff, in- lain with a compact drab or yellowish-drab clay. filtration, and permeability are slow (S), This soil is common in lowlands in eastern BEAUMONT CLAY.—Beaumont clay soil is acid, Arkansas. Calhoun silt loam is low in total nitro- poorly drained, and very slowly permeable (29). gen and phosphorus, and it is acid to strongly It occurs mainly in southwest Louisiana and in acid. Rice is commonly grown on Calhoun silt Texas, east of the Trinity River. loam in the "North-end" of the Arkansas rice area. SHARKEY CLAY.—The surface soil of Sharkey Texas clay is a dark, drab, or grayish-brown silty clay In addition to Beaumont clay soil, which com- usually mottled with brown. This is underlain at prises about one-fourth of the riceland in Texas, varying depths with a drab, steel-gray or blue, rice is grown on Lake Charles clay, Bernard sticky clay. Sand is frequently found in the surface clay loam, Edna fine sandy loam, Hockley fine layer. Sharkey clay occurs commonly in the Mis- sandy loam, and Katy fine sandy loam (£9). sissippi River bottoms. It is known as "buckshot LAKE CHARLES CLAY.—Lake Charles clay is land" because it granulates and forms a crumb the principal heavy soil in the rice belt west of structure. It can be plowed when wet; and as it the Trinity River in Texas. It comprises prob- dries out, it breaks down into granules or into clods ably one-third to one-half of the rice acreage that are easily slaked by rain. Under natural con- around Houston, Angleton, Bay City, and El 488-871 O—73^—6 78 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

Campo. Lake Charles clay is darker and more transported material of mixed geological origin. granular than Beaumont clay. It is slightly acid It was developed under poorly drained, marshy to mildly alkaline in reaction, with a pH of 6 to 8. conditions. The dark-gray or dark, brownish- BERNARD CLAY LOAM.—Bernard clay loam is gray surface soil, 14 to 24 inches deep, is coarse similar to Lake Charles clay but is more loamy and lumpy. It is heavy textured, and contains and slightly less dark, and it occurs on slightly varying quantities of completely or partly de- higher elevations. It is found both east and west cayed organic matter. The transition to a vari- of the Trinity Eiver in Texas. This soil is able subsoil, made up of stratified layers of min- slightly acid to neutral in reaction, with a pH eral and organic soil material, is rather abrupt. of 6 to 7. WILLOWS SOILS.—Willows soils consist of stream EDNA FINE SANDY LOAM.—Edna fine sandy sediment, usually reddish or yellowish brown or loam has a grayish, sandy surface and is under- dark brown, deposited along the courses of minor lain at a depth of 6 to 12 inches by a heavy, gray creeks or in the waters of temporary lakes, and claypan. It is found principally in the western underlain by brown to light-brown, compact and and southwestern parts of the Texas rice belt. relatively impervious subsoils. HocKLEY FINE SANDY LOAM.—^The Hockley OTHER SOILS.—Other California soils fre- soils form a narrow belt along the northern part quently used for rice production include Gene- of the rice area from Cleveland, in Liberty vera, Meyers, and Yolo of sedimentary alluvial County, westward through Hockley, Sealy, and origin ; Marvin, Merced, and Sycamore of mixed Eagle Lake to western Victoria County, Texas. alluvial origin; and San Joaquin of granitic al- These sandy loam soils are underlain by friable, luvial origin. sandy clay subsoils. They are slightly more sloping and better drained than are the Katy Chemistry of Flooded Soils soils, and they require more irrigation water than the Katy soils. In the production of rice, the benefits of flood- KATY FINE SANDY LOAM.—Katy fine sandy ing the soil are recognized wherever the crop is loam occurs in large level areas adjacent to Lake grown. Senewiratne and Mikkelsen (SO) re- Charles soils and in the flatter part between the viewed the literature and provided evidence that slightly more sloping and better drained Hockley flooding enhances foliar development, tillering, soils. and earlier flowering and increases yield of rice when compared with nonflooding irrigated cul- California ture. The superior growth of rice under flooded conditions can be attributed in part to the effects Stockton and Sacramento clay (£8) and Wil- of the aquatic environment, but the chemical lows (1^) are the principal soils used for rice characteristics of flooded soils are of major im- production in California. However, several other portance in the development of the crop. soils are also used. The most distinguishing characteristic of a STOCKTON SOILS.—Stockton adobe clay is one flooded soil is the presence of standing water of the principal soils used for rice production in during part or all of the growing season. The California. The surface soil is dark-gray or black layer of flood water, creating waterlogged con- clay, 5 to 16 inches deep. When wet, it is dense ditions, exerts profound changes in the physical, and plastic, but it shrinks in drying and develops chemical, and biological status of the soil. The large blocks separated by wide cracks. The upper immediate effect of flooding is a drastic curtail- part of the dark, grayish-brown, heavy clay sub- ment of gaseous exchange between the atmosphere soil is similar in structure and consistency to the and the soil. Water fills the soil pores, reducing surface soil but is calcareous. The subsoil has oxygen entry and often allowing accumulation of slightly more colloidal clay and less organic mat- gaseous products of anaerobic decomposition. ter than the surface soil. Carbon dioxide concentrations build up in the SACRAMENTO SOILS.—Sacramento clay soil oc- flooded soils, together with methane, hydrogen, cupies low-lying flood plains and is derived from nitrogen, and various oxides of nitrogen. RICE IN THE UNITED STATEiS 79

Entry of oxygen is not completely restricted found to decrease in pH over a similar range. Gen- but is confined largely to a thin layer of soil at erally, the greater the pH variance from pH 7.0 the soil-water junction {23). Oxygen arises the greater is the pH shift after flooding. This pH from gaseous exchange with the atmosphere and change is largely caused by biochemical processes as a product of photosynthesis of phytoplankton occurring in soils virtually devoid of oxygen. and hydrophytes. Chemically reduced soil constituents and carbon Pearsall {2If) and Pearsall and Mortimer (^5), dioxide gas production are the principal causative working with naturally flooded soils and lake factors. muds, determined that there were significant Another generally observed result of flooding differences between the soil at the soil-water junc- in the increased specific conductance of the soil tion and the soil immediately beneath. Flood- solution. The phenomenon is well established, water containing some dissolved oxygen main- but detailed information is lacking on the spe- tained a thin surface layer of soil in an oxidative cific nature of the increased composition of dis- condition with soil color characteristics and solved solids in the soil. The increase in the physico-chemical and biological properties dif- concentration of ammonium, iron, and manga- ferent from those of the soil beneath. Oxidation- nese ions, and in the bases displaced by these reduction potentials in the oxidative layer ex- ions from the soil exchange complex may in part ceeded 320 to 350 millivolts at a pH of 5.0 and account for the increased specific conductance. contained such oxidized chemical radicals as The increased composition of carbon dioxide, nitrates, sulfates, and ferric and manganic ions. especially at high pH values, would increase the The underlying soil was conspicuous by the ab- bicarbonate ion concentration. sence of oxygen and the concomitant presence of Several mechanical properties of soils, including the reduced forms of chemical radicals such as permeability, plasticity, cohesion, and consistency, ammonium, ferrous iron, and manganous manga- are changed by flooding. Ordinarily, these changes nese; nitrogen gas and its oxides; and various do not adversely influence the growth of rice. Alli- Sulfides, including hydrogen sulfide. The oxida- son {2) showed that the permeability of a flooded tion potentials were generally below 350 milli- soil is influenced by the amount of air entrapped volts at a pH of 5.0 in the reducing layer. Mik- during submergence. He noted that permeability kelsen and Finfrock {17) showed that reducing decreased after flooding but increased again as conditions in Stockton clay develop about 3 days air trapped in the soil was released. At a later after a soil is flooded. Patrick and Sturgis {23) period after flooding, permeability may again de- showed that when soils are flooded, the soil oxy- crease if bacteria-produced material blocks the gen disappears within a few hours and may be pores and restricts water movement. nearly absent at depths exceeding about one-half Sturgis {32) called attention to possible dele- inch. terious effects of flooding on soil structure. He The oxidation-reduction status of the soil under cited evidence that low productivity of some Lou- flooded conditions is governed by several factors, isiana soils Avas caused by deflocculation during including the rate of oxygen exchange, microbial flooding. The specific factors associated with the activity, the soil content of decomposable organic deflocculation were not reported. matter, and the base saturation status. In addition to other effects, flooding has an Flooding temporarily changes the soil pH. The important influence on soil temperature and con- change depends partly on the initial pH value sequently affects the growth of rice both directly and organic matter content of the soil and partly and indirectly. The high specific heat and high on the period of submergence. The pH change heat of vaporization of water prevent its tem- varies usually between 0.5 and 2.0 pH units. The perature from dropping as fast or as low as air tendency exists for soils to adjust to a pH ap- temperature. Under some circumstances, this modi- proaching neutrality. Eeed and Sturgis (;27) fication of the plant environment may enhance the showed that in acid soils, pH increased from growth of rice. 0.82 to 1.55 pH units, depending on soil conditions Upland soils with good aeration usually main- and initial pH values. Alkaline soils have been tain diverse populations of microflora and micro- 80 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

fauna. These organisms decompose organic matter tion. In flooded soils, the mineralization process from which they derive energy and liberate car- produces ammonium ions. The number of ammo- bon dioxide. They exert a heavy demand for nium ions reaches a plateau value rather rapidly nutrients, especially nitrogen, as they produce and tlien gradually declines. cellular material. Under flooded conditions, the The occurrence of two distinct layers—an oxi- normal soil complement of actinomycetes, fungi, dative layer at soil-water junction and a reduc- bacteria, algae, and protozoa is modified to pro- tion layer immediately beneath—exerts a signifi- duce a microflora consisting principally of anaer- cant influence on agronomic factors associated with obic bacteria, some facultative anaerobes, and a rice production. These layers should be taken into modified population of algae. The anaerobic forms account in seedbed preparation, crop residue man- have a much lower energy requirement than the agement, fertilization, cultivation, and water aerobic forms and consequently decompose organic management. matter at a slower rate. The slower decomposition Two types of nitrogen transformations occur in of organic matter under flooded conditions is flooded soils, depending on whether oxidizing or partly responsible for the development of peat in reducing conditions prevail. In the thin oxidizing bogs and marshes. The modification in the soil layer at the surface, the nitrogen transformations microflora and microfauna is only temporary, and are similar to those that occur in well-drained soils. normal populations develop when drainage and Organic matter in this layer is mineralized to am- aeration are maintained. monium ions and ultimately to nitrate ions by the The decomposition of organic matter proceeds actions of highly specialized autotropic micro- at a slower rate in flooded soils than in upland soils, organisms. The nitrate ions may be used by plants and the end products are different. Tenney and or they may be immobilized by micro-organisms, Waksman {34) showed that the decomposition of depending on the carbon-nitrogen ratio of the or- cornstalks under flooded conditions proceeds at ganic matter; or they may be moved into the about one-half the usual rate; and with more re- underlying reducing zone by leaching. The nitro- sistant materials, high in lignin, decomposition gen status of the reducing zone is characterized under flooded conditions is delayed to a much by the denitrification of nitrates and the accumula- greater extent. In well-drained soils, the end prod- tion of ammonium ions. The ammonium ions that ucts of organic matter decomposition are princi- are produced by the more sluggish bacterial micro- pally carbon dioxide, nitrates, and sulphates. flora are not reduced and provide the reservoir of Under flooded conditions, the end products include nitrogen for use by the rice crop. Nitrates that methane, hydrogen, various organic acids, ammo- move into or originate in this layer are reduced to nium ions, nitrogen and its various oxides, amines, nitrites and finally to nitrogen gas or its oxides. mercaptans, and hydrogen sulfide. These gases ultimiately escape into the air. This The rate of organic matter decomposition in soils reduction of nitrates is favored by low oxygen depends on the kind of organic constituents and tensions and the presence of oxidizable organic their nitrogen content and the carbon-nitrogen matter. Janssen and Metzger {10) have amply ratio. Ordinarily, if nitrogen is adequate for micro- demonstrated that nitrate nitrogen accumulates in bial function, nitrogen is mineralized ; and if car- well-drained soils, in contrast to flooded soils where bonaceous materials are in excess, nitrogen is ammonium ions accumulate. immobilized. Data from various sources indicate The practical significance of the differentiation that organic matter decomposition under anaerobic of distinct oxidation and reduction layers is ap- conditions proceeds at lower total nitrogen values parent in results demonstrating the poor efficiency than the 1.2- to 1.5-percent nitrogen values estab- of nitrate-nitrogen fertilizer sources in continu- lished under aerobic conditions. California field ously flooded soil and the desirability of fertilizer studies indicate that the nitrogen requirements placement of ammonium nitrogen sources {20)» for decomposition of rice straw in a flooded soil Nitrate nitrogen that is applied as a basal ferti- is as low a 0.54 percent nitrogen {SS). lizer or that develops in the oxidation layer is In well-aerated soils, the mineralization of largely lost through leaching and subsequent deni- organic matter gradually increases nitrate produc- trification. Since reducing conditions develop after RICE IN THE UNITED STATEiS 81 flooding, drilling ammonium nitrogen several tion status, the kind and amount of organic matter, inches into the soil before flooding provides good and microbial activity. Manganese occurs in three retention and availability of the nitrogen for the valence forms, with some compounds containing rice crop {17). manganese in two valence forms. In some respects, Flooding a soil often provides conditions for the the behavior of manganese in the soil is similar to increased availability of the native phosphorus. that of iron. In flooded soils, the higher oxides of Evidence of this is obtained in both the enhanced manganese are reduced to soluble and exchangeable uptake of phosphorus by rice and the increased ions. Biological reduction occurs independent of solubility of phosphorus, as shown by soil-test ex- soil pH if low oxidation-reduction potentials exist. traction methods {13, 20, SI). Factors that appear Decomposing organic matter intensifies manganese to be associated with increased phosphorus availa- reduction, especially in the lower soil pH range. bility include pH modification, reduction of insolu- It is unusual for soil manganese in flooded soils to ble ferric phosphate to the more soluble ferrous affect growth of rice adversely. High levels of sol- form, hydration, and subsequently increased hy- uble and exchangeable manganese in the soil may drolysis of ferric and aluminum phosphates. be toxic to sensitive crops grown in rotation after Sulfur in organic forms or as the sulfate ion may rice. become reduced to sulfides in flooded soils. Sulfate Fertilizers is reduced by anaerobic bacteria that are active Southern Rice Area over a wide range of soil pH and operate under The proper use of fertilizers on rice increases low oxidation-reduction potentials. Hydrogen Sul- yield from 30 to 50 percent in the southern rice fide, the end product of bacterial reduction, may area. Practically the entire rice area in Louisiana occur in gases produced in flooded soils and may and Texas requires additions of commercial fer- be present in amounts toxic to rice. This occasion- tilizers for economical yields. This is also true ally occurs in light-textured soils under extreme with most of the rice soils in Arkansas, particu- reducing conditions and in the presence of large larly in the traditional ricegrowing area of the amounts of readily decomposable organic matter. Grand Prairie region; however, the newer rice Normally, soils contain sufficient active iron to com- bottomland soils on the delta areas of Arkansas pletely precipitate the sulfide ion as ferrous sulfide and Mississippi may not require fertilizer during and prevent sulfides from causing plant damage. the first year or two of rice cropping. Iron, a prominent constituent of the soil, occurs Commercial fertilizers were not used extensively in primary minerals, hydrated oxides, silicate in the southern rice area before World War II. In clays, and various organic complexes. When a soil fact, results from may of the soil fertility tests con- is flooded, the iron undergoes considerable changes ducted in the early thirties showed negative yield in solubility, which is greatly influenced by anaero- responses from fertilizer applications {W). This bic bacteria. The extent of the change is a function was usually true when various rates and ratios of of the organic matter content, the low^ oxidation- nitrogen and phosphorus had been applied at seed- reduction potential, and the soil reaction. Ferric ing. This method of application greatly stimulated iron, which predominates in well-drained soils, is weed and grass growth before the rice plants be- reduced to the ferrous form, especially as hydrox- came established, and this competition often se- ides and carbonates. With sustained flooding, verely reduced rice yield. However, the real worth equilibrium develops between the soil and the soil of commercial fertilizers became apparent with the solution, wâth a significant increase in soluble and gradual improvement of irrigation and drainage exchangeable iron. Both mineral and organic salts of iron appear in the soil solution. Ordinarily, the facilities, improved land preparation, mechanized abundance of soluble iron does not adversely affect harvesting equipment, the development of varieties the growth of rice but in some situations toxicities suitable for mechanization, and the greatly in- or nutrient antagonism may impair growth. creased use of the airplane for applying fertilizers. Manganese chemistry of the soil is not w^ell un- Fertilizers were also greatly improved. Pelleted derstood. Its forms are in dynamic equilibrium, or prilled, granular, and large crystalline, high- depending on such factors as pH, oxidation-reduc- analysis fertilizers are available and are ideally 82 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

suited for application with the airplane or with the rice has completed one-half of its growth improved ground equipment (i, 3). period. In Arkansas, a nitrogen application about Knowledge of previous cropping history and a 60 to 70 days before harvesting has produced good soil test are useful in determining the amount and results. Thus, the timing depends to a large extent kind of fertilizer to use on rice in some Southern on the total growth period of the particular rice States (^, 29). Specific fertilizer recommenda- variety. tions vary considerably between States and from The rice soils in Louisiana and Texas are gen- area to area within States, or even from farm to erally deficient in nitrogen and phosphorus. Thus, farm within the same area. The rice variety, the addition of 80 to 120 pounds of nitrogen per water management, methods of weed control, and acre and 20 to 40 pounds of phosphoric acid other managerial variables affect use of fertilizers (P2O5) are usually necessary for economical rice (iJ). Various rapid, chemical, soil-testing meth- production {36). In most instances, no potash is ods are used effectively to determine the fertilizer required. The small areas of sandy soils in both and lime requirements of rice soils (4). Chemical States, however, sometimes respond to a 20- to determination can be made on soil reaction (pH), 40-pound per acre application of potash. percentage of organic matter, and the available When applying fertilizer as a topdressing on phosphorus, potassium, and calcium; and the rice growing in heavy soils, placing it on dry soils salinity hazard can be noted. The nitrogen level is usually preferred to placing it on wet or flooded is usually estimated from the organic matter soils. Soon after the fertilizer is applied, the content. Results of these tests obtained from air- fields are flooded; thus the water is an effective dried soil samples can be interpreted with reason- carrier to move the fertilizer into the root area. able accuracy, provided basic information is If scarcity of irrigation water, weed infestations, available on the complex chemical changes known rains, or timing difficulties prevent flooding, the to occur under submergence. All this information rate of application, particularly of nitrogen, is is of value in determining fertilizer requirements increased slightly to compensate for reduced effi- for rice in southern rice areas. ciency. The rate for nitrogen should be increased Nitrogen is used at somewhat higher rates in about 5 to 10 percent on wet soils and perhaps Arkansas than in the other States. Rates of more 10 to 15 percent on flooded soils {'29). than 100 pounds per acre of actual nitrogen are Applying a large amount of fertilizer directly rather common, particularly on the soils of the with the seed is hazardous, because germination Grand Prairie {36). Some differences in the ferti- may be reduced or emergence may be delayed. The lizer requirements of rice varieties are recognized, time of seedling emergence directly influences with the stiffer strawed varieties being capable of the flooding date and thus becomes important in using higher rates of nitrogen than lodging-sus- controlling grass weeds. These problems do not ceptible varieties. Phosphorus and potash are usu- ordinarily occur if the fertilizer is placed 2 to 3 ally applied on the basis of a soil test. The nitrogen inches below the seeds. requirements for rice growth in the delta areas of Limited reseai'ch on the mineral deficiency Arkansas and Mississippi seldom exceed 60 to 80 symptoms of rice has been conducted in the pounds per acre. These newer soils frequently United States. Olsen {22) reported typical foliar require no fertilizer during the first year or two symptoms due to deficiencies of nitrogen, phos- of rice production. phorus, potassium, calcium, magnesium, and iron The timing of fertilizer application on rice is in three greenhouse experiments. A deficiency of very important. Most rice farmers in Louisiana each of these elements, except calcium, reduced and Texas prefer a somewhat earlier application tillering. Reduced tillering caused by nitrogen of the total amount of fertilizer than farmers in deficiency was particularly pronounced. All ele- Arkansas and Mississippi {35). For all States, ments reduced root and top development; potas- however, all of the phosphorus and potassium and sium and nitrogen deficiencies caused the most part of the nitrogen should be applied as near to severe reductions. seeding time as possible. In Louisiana and Texas, The sources of nitrogen, phosphorus, and po- the remaining nitrogen should be applied before tassium used in southern rice areas vary widely RICE IN THE UNITED STATEiS 83 and to a large extent depend on the cost of ap- usually about three-fourths of the amount applied plication and physical condition of the fertilizer. to the first crop, should be applied immediately Ammoniacal forms of nitrogen are generally after first harvest. This practice consistently gave preferred to the nitrate forms because nitrates are rice yields that were one-third to one-half as much difficult to maintain in a flooded soil. Such am- as the original crop. Ordinarily, fertilizers con- moniacal materials as ammonium sulfate, am- taining phosphorus and potassium need not be monium phosphate, urea, and ammonium chloride applied, because residual quantities of these ele- give about equally good results. Because of the ments applied to the first crop fulfill the ratoon lower cost of application, there is a deñnite trend requirements. toward the use of high-analysis fertilizers. Anhy- drous ammonia is a good source of nitrogen for California rice, although it is difficult to apply with ground Fertilizers, particularly nitrogen, increased rice equipment during wet periods. Application in yields in the earliest experiments conducted in water is satisfactory only when precise watering California, in 1914-16 {11). Dunshee (7) con- methods are used, because uniform distribution of tinued rice fertilizer research, and it was greatly the material depends on having a uniform dis- expanded by Davis and Jones {6) from 1925 tribution and depth of water. through 1937. Davis and Jones showed that nitro- Phosphorus is usually supplied in the form of gen fertilizers improved yields significantly on superphosphate, ammonium phosphate, or diam- Stockton clay adobe. Applications at the time of monium phosphate. Rock phosphate is sometimes seeding were more effective than were later appli- used and is satisfactory. cations. They also reported that phosphate and Potassium chloride is the most common source potassium fertilizers did not increase yields on of potassium. Only limited quantities of potas- Stockton clay adobe. sium sulfate are used, although it is considered More recent fertilizer research has demonstrated equal to potassium chloride in its benefits to the the need for nitrogen on all soils except those on crop. which a good leguminous green manure crop is In areas where crops preceding rice have been grown {18), Phosphorus fertilizers have increased fertilized with phosphorus and potash, residual yields on the brownish-red terrace soils bordering amounts of these elements may be sufficient for the Sacramento and the San Joaquin Valleys. Rice one or perhaps two consecutive rice crops. yields on some basin soils after years of cropping Recent research with micronutrient elements are better when phosphorus is included with nitro- shows that a chlorosis in seedling rice causing gen. In some areas, usually parts of large fields, reduced plant populations and delaying maturity where alkali salts have accumulated and where the is correctable by the application of zinc com- soil reaction exceeds pH 8.5 in a 1:10 soil-water pounds. On calcareous or alkaline soils sulfur or paste, dramatic responses have been obtained from sulfuric acid may partly alleviate the problem by various impure iron sources {9), Recent results acidifying the soil {26, 37). prove that most of their benefit was due to zinc Addition of limestone has not been necessary for contained in the impure products (fig. 28). rice grown on moderately to slightly acid (pH 5.0 Rice was fertilized in California before 1953 to 6.5) soils. Of equal importance is the fact that by broadcasting either on the dry seedbed before no detrimental effects on rice yields have been flooding and planting or, more commonly, noted from adding lime, which is usually added at on the water by airplane after seeding. Ferti- the rate of 1 to 2 tons per acre on crops rotated lizer placement work of Mikkelsen and Finfrock with rice. {17) demonstrated that nitrogen broadcast on the The long growing seasons in southwestern Lou- soil surface or applied to the flooded fields was isiana and southeastern Texas permit the produc- not used efficiently by rice. Broadcast nitrogen tion of a ratoon or stubble crop, particularly by was lost through nitrification and subsequent the earliest maturing rice varieties seeded around denitrification. However, ammonium nitrogen the middle of April. Research in Texas {8) has drilled 2 to 4 inches into the soil, where reducing shown that for best results, additional nitrogen. conditions developed 3 to 5 days after flooding, 84 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

Phosphate fertilizer should be applied before flooding, usually simultaneously with basal nitro- gen application. Phosphorus does not move ap- preciably from where it is applied; but because rice roots are relatively shallow, results from placed and broadcast phosphorus are about equal. Topdressings with phosphorus can be effective on established rice if it is applied during the tiller- ing stage. The quantity of nitrogen fertilizer used in rice production varies from 30 to 120 pounds of actual nitrogen per acre. This is supplied as one of tlie commercially available ammoniacal sources such as ammonium sulfate, urea, anhydrous ammonia, or ammonium phosphate sulfate mixtures. On soil low in nitrogen, as much as 80 to 120 pounds of nitrogen per acre (400 to 600 pounds of ammo- nium sulfate) is applied. On soils of average fer- tility, producing about 50 hundredweight of paddy rice, 60 to 80 pounds of nitrogen per acre is applied. On most California rice soils, applications of PN-2785 FiGUBE 28.—zinc deficiency corrected by soil application 40 to 60 pounds of P2OS per acre will supply the of zinc fertilizers. Soil in upper part of photograph was phosphorus needs of rice. Some residual effects not treated with zinc. have been observed on subsequent crops, but the carryover from a single application may not be sufficient for best yields during a second year. In remained in the soil and was continuously avail- areas where phosphorus is needed, the added able to the rice plants. growth and yield often require that additional ni- The time of applying fertilizer before flooding trogen be supplied. AVhere rice produces better is important, because nitrification is undesirable growth with phosphorus, nitrogen rates should be both before and during flooding. Mikkelsen {16) increased 25 to 50 percent. establislied that nitrification occurs if ammonium Experiments to determine the best nitrogen nitrogen is drilled into a warm, moist, aerated sources for rice have been conducted over a long seedbed before flooding. As much as 60 percent of period. Davis and Jones {6) compared ammo- the ammonium nitrogen can be converted to ni- nium sulfate, Ammo-Phos, Leunasalpeter, urea, trate in 7 days of typical spring soil temperatures. Ammo Phos Ko, Leunaphos, Calurea, and cyan- Split application of nitrogen, with part placed amide during 1932-36. They concluded that in the soil before flooding and the rest used as a ammonium sulfate was the most profitable. Ex- topdressing during the period of panicle forma- periments in which nitrogen from different tion, has proved effective in some regions of the sources was drilled into the soil before flooding world. Topdressing experiments in California are reported by Mikkelsen and Miller {19). In rice production have shown no superiority over yield comparisons with ammonium sulfate rated preplant soil application {19). Where application as 100, ammonium chloride ranked 97, cyanamide of nitrogen to the seedbed was not sufficient to 92, urea 90, aqua ammonia 85, anhydrous am- maintain normal color and growth of rice, sup- monia 83, and ammonium nitrate 57. Aqua am- plemental applications have been profitable. For monia and anhydrous ammonia are good nitrogen effective use of topdressed nitrogen on California sources but in loose dry seedbeds they sometimes rice varieties, the application should be made no do not perform as well as dry materials because later than the jointing stage of growth. of volatization losses. RICE IN THE UNITED STATEiS 85

Various diagnostic techniques have been stud- above or below the critical range. If the nutrient ied for evaluating the fertilizer needs of rice. A concentrations of each element are above the crit- method of plant analysis based on the principle ical range until at least 80 days after sowing Cali- of limiting factors and critical nutrient concen- fornia rice varieties, benefits from additional trations to denote probable levels of nutrient de- amounts of this nutrient are not likely to occur. ficiency is applicable to rice. Through comparisons However, if the nutrient concentration of any of nutrient concentrations of plants restricted in element falls below the critical concentration, the growth and yield with those not restricted, "crit- proibability of a response from fertilization is very ical nutrient concentrations" for nitrogen, phos- high. phorus, and potassium have been developed. The The "critical nutrient concentrations" estab- probability of obtaining a yield response from lished for California rice varieties, using the most the addition of a nutrient thus depends upon recently matured leaf as the test material are as whether the critical nutrient concentration is follows :

Rice Plant Analysis Guide"^

Kjeldahl nitrogen Phosphate-phosphorus Potassium Growth state Critical Adequate Critical Adequate Critical Adequate concentration range concentration range concentration range

Percent Parts per million Percent Mid-tillerting 3. 0 3. 0-4. 0 1, 000 1, 000-1, 800 1.2 1. 4-2. 8 Maximum tillering 2. 6 3. 8-3. 6 800 1, 000-1, 800 1. 0 1. 2-2. 4 Panicle initiation 2. 4 2. 6-3. 2 800 1, 000-1, 800 0. 8 1. 0-2. 2

* Analysis on a dry weight basis. Kjeldahl nitrogen: 2 percent acetic acid extractable phosphate-phosphorus and potassium.

Mícronutríent Deficiencies yields vary, depending upon the severity of the deficiency and measures taken to correct them, but Deficiencies of two micronutrient elements, zinc sometimes crops fail completely. and iron, have been recognized in several of the Iron deficiencies are more closely correlated with major rice areas during the past few years (P, 26, high soil pH and exchangeable calcium than other 37). These deficiencies are not widespread but are soil characteristics already evaluated {26). Calcar- very significant where they do occur because they eous and high sodium soils are usually most suscep- inñuence stand establishment and affect the ma- tible to iron deficiencies. Iron deficiency chlorosis turity and yield of the rice crop. may be further aggravated by high levels of phos- Iron and the more frequently found zinc de- phate and bicarbonate ions either in the soil or in ficiencies are usually evident as soon as the seedling the water used to ñood rice. Iron deficiencies have rice becomes established. The young plants are been effectively corrected by the use of various iron usually structurally weak when they emerge from salts such aö ferrous sulfate, ferrous ammonium the soil and become chlorotic by the time the first sulfate, iron oxides, and ferric sulfate. Iron de- true leaves are formed. The intensity of yellowing ficiency has also been corrected by the use of acidu- usually increases and ultimately irregular, brown, necrotic spots develop on the leaves. In some in- lation of the floodwater with sulfuric acid. stances affected plants will die within a few weeks Zinc deficiency in rice occurs most frequently in after planting. Sometimes if plants can survive soils where the pH may range from neutral to this early condition, they may resume growth and strongly alkaline. It is sometimes associated with ultimately regain their normal color and produce soils of high organic matter content and occurs grain, although maturity may be delayed. Grain frequently in areas where the topsoil has been 86 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

removed during land leveling. The combined ef- (9) INGEBRETSEN, K., MARTIN, W. E., VLAMIS, JAMES, fects of high pH, rapid organic matter decomposi- and JETER, ROY. tion releasing carbon dioxide, and a calcareous soil 1959. IRON DEFICIENCY OF RICE. Calif. Agr. 13: 6-7, 8, and 14. condition often aggravate zinc deficiency. High (10) JANSSEN, GEORGE, and METZGER, W. H. levels of soil phosphorus also tend to increase the 1928. TRANSFORMATION OF NITROGEN IN RICE SOIL. severity of zinc deficiency. Amer. Soc. Agron. Jour. 20: 459^76. Soluble zinc salts such as zinc sulfate and even (11) JONES, J. W. sparingly soluble zinc oxide have corrected zinc 1923. RICE EXPERIMENTS AT THE BIGGS RICE FIELD STATION IN CALIFORNIA. U.S. Dept. Agr. deficiency (fig. 28). Chelates also correct zinc de- Dept. Bui. 1155, 60 pp., iUus. ficiency but have not performed as well as the (12) DocKiNS,, J. O., WALKER, R. K., and DAVIS, metallic salts. Applications of zinc materials be- W. C. fore planting have usually given the best results. 1952. RICE PRODUCTION IN THE SOUTHERN STATES. U.S. Dept. Agr. Farmers' Bui. 2043, 36 pp., Applications made as a topdressing to the crop can inus. be effective, but the effectiveness of the treatment (13) MCGREGOR, A. J. diminishes with time. Foliar applications of zinc 1953. PHOSPHATE MOVEMENT AND NATURAL DRAIN- solutions can be made, but the limited leaf area of AGE. Jour. Soil Sei. 4 : 86-97. (14) MANN, C. W., WARNER, J. F., WESTOVER, H. L., and seedling rice and the problem of wetting the leaves FERGUSON, J. E. greatly limit its usefulness. Under most conditions 1911. SOIL SURVEY OF THE WOODLAND AREA, CALI- the application of 8 to 12 pounds of zinc per acre FORNIA. U.S. Dept. Agr., Bur. Soils, Soil Survey Adv. Sheet 1909, 57 pp. as a zinc salt adequately corrects this deficiency. (15) MiEARS, R. J. 1958. FERTILIZER AND CULTURE PROBLEMS. Rlce Selecfed References Tech. Working Group Proc. 8: 17-19. (16) MiKKELSEN, D. S. (1) ADAIR, C. R., and ENGLER, KYLE. 1962. NITROGEN FERTILIZATION OF JAPÓNICA RICE 1955. THE IRRIGATION AND CULTURE OF RICE. In IN CALIFORNIA. Rice Jour. 65(3) : 8-13. Water, U.S. Dept. Agr. Yearbook of Agr., (17) and FINFROCK, D. C. pp. 38i>-394. 1957. AVAILABILITY OF AMMONIACAL NITROGEN TO (2) ALLISON, L. E. LOVTLAND RICE AS INFLUENCED BY FERTILIZER 1947. EFFECT OF MICROORGANISMS ON PERMEABILITY PLACEMENT. Agrou. Jour. 49: 296-300. OF SOIL UNDER PROLONGED SUBMERGENCE. (18) LiNDT, J. H., Jr., and MILLER, M. D. Soil Sei. 63 : 439-450. 1967. RICE FERTILIZATION. C^lif. Agr. Expt. Sta. (3) BEACHELL, H. M. Ext. Serv. Leaflet 96, rev., 12 pp., illus. 1959. RICE. In Matz, S. A., ed., The Chemistry (19) and MILLER, M. D. and Technology of Cereals as Food and 1963. NITROGEN FERTILIZATION OF RICE IN CALIFOR- Feed, pp. 137-176. Avi Pub. Co., Inc., West- NIA. Calif. Agr. 17: 9-11. port, Conn. (20) MITSUI, SHINGO. (4) BEACHER, R. L. 1954. INORGANIC NUTRITION, FERTILIZATION AND 1955. SAMPLING AND ANALYSIS OF PADDY SOILS. SOIL AMELIORATION FOR LOWLAND RICE. 107 Internatl. Rice Comn. Meeting Proc, FAO, pp. Yokendo Ltd., Tokyo. Penang, Malaya. (21) NELSON, MARTIN, SACHS, W. H., and AUSTIN, R. H. (5) CLARK, H. L., HALEY, G. J., HEBERT, E. J., and others. 1923. THE SOILS OF ARKANSAS. Ark. Agr. Expt. 1962. SOIL SURVEY OF ACADIA PARISH, LOUISIANA. Sta. Bui, 187, 83 pp., illus. U.S. Soil Conserv. Serv. in coop, with La. (22) OLSEN, K. L. Agr. Expt. Sta. Ser. 1959 No. 15, 57 pp. 1958. MINERAL DEFICIENCY SYMPTOMS IN RICE. (6) DAVIS, L. L., and JONES, J. W. Ark. Agr. Expt. Sta. Bui. 605, 11 pp. 1940. FERTILIZER EXPERIMENTS VTITH RICE IN CALI- (23) PATRICK, W. H., JR., and STURGIS, M. B. FORNIA. U.S. Dept. Agr. Tech. Bui. 718. 1955. CONCENTRATION AND MOVEMENT OF OXYGEN 21 pp. AS RELATED TO ABSORPTION OF AMMONIUM (7) DUNSHEE, C. F. AND NITRATE NITROGEN BY RICE. Soil Sci. 1928. RICE EXPERIMENTS IN THE SACRAMENTO VAL- Soc. Amer. Proc. 19 : 59-62. LEY, 1922-1927. Calif. Agr. Expt. Sta. Bui. (24) PEARSALL, W. H. 454, 14 pp., illus. 1938. THE SOIL COMPLEX IN RELATION TO PLANT (8) EvATT, N. S., and BEACHELL, H. M. COMMUNITIES. I. OXIDATION-REDUCTION PO- 1962. SECOND-CROP RICE PRODUCTION IN TEXAS. TENTIALS IN SOILS. Jour. Ecol. 26: 180- Tex. Agr. Prog. 8(6) : 25-28. 193. RICE IN THE UNITED STATES 87

(25) PEARSALL, W. H., and MORTIMER, C. H. (32) STURGIS, M. B. 1939. OXIDATION-REDUCTION POTENTIALS IN WATER- 1936. CHANGES IN THE OXIDATION-REDUCTION LOGGED SOILS. NATURAL WATERS AND MUDS. EQUILIBRIUM IN SOILS AS RELATED TO THE Jour. Ecol. 27 : 483-501. PHYSICAL PROPERTIES OF THE SOILS AND THE (26) PLACE, G. A. GROWTH OF RICE. La. Agr. Expt. Sta. Bui. 1969. RELATIONSHIP OF IRON, MANGANESE AND BI- 271, 37 pp. CARBONATE TO CHLOROSIS OF RICE GROWN IN (33) CALCAREOUS SOILS. Ark. Expt. Sta. Rpt. 1957. MANAGING SOILS FOR RICE. In SoilS, U.S. Ser. 175, 21 pp. Dept. Agr. Yearbook of Agr., pp. 658-663. (27) REED, J. F., and STURGIS, M. B. (34) TENNEY, F. G., and WAKSMAN, S. A. 1939. CHEMICAL CHARACTERISTICS OF THE RICE 1930. COMPOSITION OF NATURAL ORGANIC MATERIALS AREA OF LOUISIANA. La. Agr. Expt. Sta. AND THEIR DECOMPOSITION IN THE SOIL. V. Bui. 307, 31 pp. DECOMPOSITION OF VARIOUS CHEMICAL CON- (28) RETZER, J. L., GLASSEY, T. W., GOFF, A. M., and STITUENTS IN PLANT MATERIALS, UNDER HARRADINE, F. F. ANAEROBIC CONDITIONS. Soll Sci. 30: 143- 1951. SOIL SURVEY OF THE STOCKTON AREA, CALI- 160. FORNIA. U.S. Dept. Agr. in coop, with Univ. (35) THOMPSON, L., MAPLES, R., WELLS, J. and others. Calif. Agr. Expt. Sta. Ser. 1939, No. 10, 1962. RECOMMENDATIONS FOR RICE FERTILIZATION IN 121 pp. SOUTHERN STATES. Rice Jour. 65(1) : 5, 6, (29) REYNOLDS, E. B. and 40. 1954. RESEARCH ON RICE PRODUCTION IN TEXAS. (36) WALKER, R. K., and MIEARS, R. J. Tex. Agr. Expt. Sta. Bui. 775, 29 pp. 1957. THE COASTAL PRAIRIES. In Soil, U.S. Dept. (30) SENEWITARNE, S. T., and MIKKELSEN, D. S. Agr. Yearbook of Agr., pp. 531-534. 1961. PHYSIOLOGICAL FACTORS LIMITING GROWTH (37) WESTFALL, D. G., ANDERSON, W. B., and HODGES, R. J. AND YIELD OF ORYZA SATIVA UNDER FLOODED 1969. MICRONUTRIENTS AND RICE PRODUCTION. CONDITIONS. Plant and Soil 14: 127-146. Rice Jour. 72: 67-68. (31) SHAPIRO, R.E. 1954. THE EFFECT OF FLOODING ON AVAILABILITY OF (38) WILLIAMS, W. A., MIKKELSEN, D. S., MUELLER, K. E., SOIL PHOSPHORUS YIELD, AND PHOSPHORUS and RUCKMAN, J. E. AND NITROGEN UPTAKE BY RICE. FOUrth lu- 1968. NITROGEN IMMOBILIZATION BY RICE STRAW natl. Rice Gomn. Proc, Tokyo [Mimeo- INCORPORATED IN LOWLAND RICE PRODUCTION. graphed.] Plant and Soil 28 : 49-60. CULTURE

By T. H. JOHNSTON and M. D. MILLER

Eice has been grown as a commercial crop in Rotation or Cropping Systems the United States since the latter part of the 17th century. Eice cultural methods used from In most rice-pi'oducing areas of the United that time until the present have been reviewed States, crops are rotated because under continu- by Adair, Miller, and Beachell (8). They traced ous cropping the soil usually becomes depleted the evolution of cultural methods from the use in fertility and in organic matter. The resulting of hand labor for clearing timber from the deterioration of the physical condition of the soil land, for digging canals and ditches, for plow- makes seedbed preparation especially difficult. In ing with a hoe, for seeding, harvesting, and addition, the soil usually becomes progressively threshing, through the use of animal power infested with weeds and diseases that lower the (oxen, mules, and horses) for binding and thresh- yield and quality of the rice. ing the rice. Then came steam-powered threshers. In the early years in the Carolinas, rice was Today, several large "rice-special" diesel- and gas- grown continuously in the same field with only powered tractors and self-propelled combines occasional rest (45). Later, ricefields in that area may be used on one farm. Much of the seed- sometimes were planted to oats in the fall, followed ing and most of the fertilizing and spraying by potatoes the next year. Some farmers grew rice is now done by airplanes that can cover several and cotton in alternate years. This helped to con- hundred acres a day. trol w^eeds in both crops. Much of the increase in rice production per In the early years in the South Central States, acre can be attributed to improved cultural meth- fields were cropped to rice year after year until the ods made possible by the invention and manufac- rice yields became low and the quality poor because ture of the specialized equipment. The major rice of the mixtures of weed seed and red rice in the areas of the United States are now described as threshed grain. Fields then were allowed to lie the most highly mechanized farming areas in the idle for 1 or 2 years, and then were again put back world. Stout estimates that as few as 7.5 hours are into rice. This helped to increase rice yields but did required annually to produce an acre of rice in not control the weeds satisfactorily. Therefore, the United States under complete mechanization, ricefields were grazed during the years that they whereas in many parts of the world from 500 to were "laid-out." This practice helped to control 800 are required {U6). grass and weeds but did not control red rice. Some Along with improvements in machinery and farmers practiced summer-fallowing for a year or farming methods have come other innovations, two between rice ciops and in this way controlled including use of reservoirs for improved water weeds and red rice more effectively than when the supply and, more recently, underground pipelines fields were idle. to eliminate many open canals. Modem riceland cropping systems are based on Eotation or cropping systems; land leveling information gained from controlled experiments and seedbed preparation; seed and seeding; ir- and from grower experience. The preferred system rigating; and harvesting, drying, and storing for any farm depends on soil type, local climatic methods have been developed and improved conditions, and economic considerations. through research. Many branches of science have Arkansas contributed to the development and adaptation of new and improved methods, equipment, and Eotation experimients were begun in 1927 at the facilities now used in rice culture. Eice Branch Experiment Station, Stuttgart, Ark. RICE IN THE UNITED STATEiS 89

Because of the numerous factors known to affect the land idle, a rotation of rice, oats, and lespedeza, rice yields, many rotations were tried, along with and a rotation of rice and soybeans. Eeturns to the several combinations of delayed seeding, tillage for operators were slightly higher with only one-third weed control, and summer-fallowing. Kesults in- of the land in rice than with one-half of the land cluded the following {98, 99) : in rice, if other crops such as oats, lespedeza, and (1) In 2-, 3-, and 4-year rotations, best yields soybeans were grown and harvested to supplement were obtained when rice was grown not more than income. half the time. Forty percent of the operators of large rice (2) Eice rotated with fallow or early-planted farms followed a cropping system including rice soybeans (for beans or hay) orlespedeza (for hay) and oats, with the oats in many cases being over- produced 1,000 pounds more rice per acre in the seeded with lespedeza; and almost all operators year it was grown than when rice was grown con- had some fallow and idle land. Approximately tinuously. Over a 7-year period of continuous crop- 25 percent of the farmers produced soybeans, and ping, rice yields declined 180 pounds per acre. about 35 percent produced beef cattle. The sec- (3) The best 3-year rotation was 1 year of soy- ond most common rotation was rice-oats-lespe- beans, followed by a winter vetch cover crop, and deza-soybeans, with 42 percent of the land beirg 2 years of rice. Rice yield increase was greatest the used for rice under this cropping system. first year. In 1970, the crops most often grown in rota- (4) As compared with continuous rice, 4-year tion with rice in Arkansas were soybeans and rotations of soybeans-oats-rice gave large yield in- oats. Legume green manure crops were used in- creases in the first year of rice, but only half as frequently. In recent years, 1 or 2 years of les- large an increase in the second year of rice. pedeza in a rice rotation has sometimes led to (5) Eice yields were significantly increased by considerable damage to rice from the so-called plowing down legume green manure crops, includ- lespedeza worm, grape colaspis (Maecolaspis ing soybeans, lespedeza, and hairy vetch (Vicia -ßavida (Say)). However, chemical controls for villosa Eoth) immediately preceding the rice this insect have been developed by Eolston and crops. All green manure treatments increased rice ^ovi^^ (115,116). yield more than did chemical fertilizer applied to The rotation of rice with reservoirs used for fish the preceding crop in the rotation. production and as a source of irrigation water has Simmons (l^i) reviewed Arkansas rice rotation been practiced in Arkansas (^7). In some in- research, discussing 3-, 5-, 6-, and 8-year systems. stances, the rotation has produced substantial in- These systems involved rice on the land for half of creases in rice yields, even without the use of the time or less and soybeans, oats, lespedeza, or commercial fertilizer on the rice crops. The rota- hairy vetch the rest of the time. tion may include 2 years of fish and 2 years of rice Perkins and Lund (108) listed ways in which or 1 year of fish and 1 or 2 years of rice. Leaving a good rotations would benefit the farmer, but cau- reservoir in fish for longer than 2 years is usually tioned that these rotations would not replace min- unsatisfactory because the accumulated fertility eral plant food elements such as phosphorus and results in excessive vegetative growth of the rice the potassium. They stated that legumes may supply first year after fish. In some cases, even 2 years in considerable nitrogen, but more nitrogen may be fish results in excessive vegetative growth and se- needed. They also stated that rotations will aid in vere lodging of the following rice crop. disease and insect control, but other control meas- Experiments by Sims (127, 128) indicate that ures may be required. a large part of the increased vegetative growth In 1947 the order of frequency of crops in the of rice may be attributed to the accumulation of Arkansas rice area was rice, oats, lespedeza, soy- ammonium nitrogen in the soil during the period beans, corn, and cotton (ISl), The most prevalent the reservoirs were in water and fish. Excessive cropping systems on small rice farms in Arkansas vegetative growth of rice may be avoided by were 4 or 6 years long, with rice being grown on growing a row crop such as soybeans, grain the land for 2 or 3 consecutive years (95, 96) : sorghum, or corn in the rotation the first year These systems included growing rice and leaving following fish or water and then growing rice 90 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE, the second year. In one field test on a clay soil erties, according to Sims (126). He found signifi- that had been in fish 2 years, all of these crops re- cantly greater amounts of available nitrogen in duced the ammonium nitrogen content of the samples from a year-old reservoir that in those soil from 175 pounds per acre at seeding time to from a "year-old" ricefield. He concluded that sub- 40 pounds per acre at harvest. Kice grown on merging a reservoir soil for 2 or more years would the experimental area the second year did not result in the accumiulation of sufficient nitrogen to lodge. meet the entire needs of the following rice crop on In ricefields where high soil fertility resulted some soils and provide an excess of nitrogen on in excessive early vegetative growth, draining other soils. the fields and allowing them to dry thoroughly Sullivan (14^) reported that some Arkansas before the development of the rice panicles farmers in recent years have started to rotate (heads) at the early jointing stage of growth water and crops. Their fields are kept flooded 1 helped retard later vegetative growth of the rice or 2 years and then seeded to rice. According to and helped reduce lodging. Sullivan, benefits under this system include in- Green (4^6) reviewed the problems of fish farm- creased organic matter content of the soil, im- ing and stated that haphazard raising of fish proved physical soil characteristics, improved must be supplanted by more scientific production weed control, and improved recreational and and marketing practices. wildlife facilities. Green and White (48) compared three selected 4-year rice rotations in eastern Arkansas. These Louisiana rotations were fish-fish-rice-rice, soybeans-soy- Eeporting on early experiments in rice pro- beans-rice-rice, and idle-fallow-rice-rice. Combi- duction in southwest Louisiana, Chambliss and nations of buffalo and bass were the fish species Jenkins (19) in 1925 stated : "Good drainage, good most commonly stocked. These studies indicated tillage, and proper rotation make unnecessary the that a total of 881 pounds of buffalo and 181 application of any commercial fertilizer to the pounds of bass per acre must be produced and mar- Crowley silt loam at the present time." Results keted during the 2-year fish period for this system obtained at the Crowley Rice Experiment Station of land management to be as profitable as the soy- from 1913 to 1923. inclusive, showed that rice in bean-rice rotation. This level of production is con- rotation with soybeans averaged 2,384 pounds per siderably higher than that reported by the 35 fish- acre, as compared with 1,243 pounds per acre for rice farmers whose operations were included in continuous rice. Efficient drainage and good till- the study. age, supplemented by the organic matter added to Preliminary experiments conducted at the Eice the soil by plowing under mature soybean plant Experiment Station at Crowley, La., showed that remnants after harvest, gave greater returns than good-quality catfish could be produced under w^ere obtained from commercial fertilizer applied conditions similar to those in flooded ricefields to the rice crop. In addition, the soil was left in a without major disease or parasite problems loose, friable condition, which facilitated the prep- (139). To help answer some of the complex aration of a better rice seedbed. problems encountered in the growing of fish and Jenkins and Jones (65) in 1944 pointed out in the use of fish-rice rotations, a Fish Farming that the rice crop, like other cereals, responds to Experimental Station was established in 1961 at appropriate cultural methods and rotation sys- Stuttgart, Ark., by the Fish and Wildlife Serv- tems. At the time their experiments were started ice of the U.S. Department of the Interior. In in 1934, the rice crop normally was grown in cooperation with the University of Arkanses, ex- alternate years or once in 3 years on land fal- periments have been initiated to determine suit- lowed or left in stubble pasture for 1 or 2 years. able management and fertilization practices In 2-year rotations, the highest yields of rice were obtained following Italian ryegrass, clov- Detailed chemical, physical, and biological anal- ers, or stubble pasture. Another experiment yses of pond soil made at the start of the study and demonstrated how other crops grown in the ro- after 1 year showed no major changes in soil prop- tation may influence rice yield. The average yield RICE IN THE UNITED STATES 91 of rice following cotton that had been dusted with yields from unimproved pastures by more than calcium arsenate was 30 percent below the yield 150 pounds per acre. In addition, turning under following cotton that had not been dusted. The improved-pasture sod ahead of rice crops increased reduced yield apparently was due to the adverse rice yields 1,000 to 1,800 pounds per acre. residual effect of the calcium arsenate upon the Black and Walker {16) reported in 1955 on a ensuing rice crbp. Reed and Sturgis {112) in 1936 5-year rotation experiment that was established showed that arsenic toxicity symptoms in the rice at five locations in Louisiana, from 1946 to 1953, plants are similar to those of straighthead in which on four soil types. The results of these five ex- case florets may be distorted and seed set may be periments substantiated earlier work on pasture- reduced markedly. rice rotation in southwest Louisiana. They found The average yields of cotton and soybeans in that if improved pastures are established, it is a 3-year rotation of cotton, soybeans, and rice desirable to provide contour levees and irriga- were too low to be profitable, and the average tion during dry periods, so that full benefits can yield of rice was slightly less from this rotation be received from the relatively large investment than from the better 2-year rotations. required to establish improved pastures. Also, it In 4-year rotations consisting of rice 2 years was necessary to obtain a minimum of 3 years' followed either by 2 years of cotton (fertilized grazing from an improved pasture before plant- and not fertilized) or by 2 years of native pas- ing it to rice because an improved pasture produces ture (fertilized and not fertilized), the yields of only about two-thirds as much in the first year rice were not increased by fertilizing ; but the rice as in the second and third years, and the initial following native pasture yielded somewhat more cost of an improved pasture is fairly high. Black than did the rice following cotton. and Walker concluded that in southwest Louisi- In 10-year rotations of five successive rice crops ana a long-time rotation of improved pasture following 5 years each in (1) improved pasture, and rice is superior to the more common rotation (2) native pasture, (3) corn plus soybeans, or of 1 year of rice followed by 1 year of native (4) cotton, the 4-year average yield per acre of pasture. rice was 2,192 pounds following improved pas- Davis, Sonnier, and White {24) in 1963 de- ture ; 2,120 pounds following native pasture ; 2,052 scribed an experiment initiated in 1953 to deter- pounds following corn plus soybeans; and 1,444 mine the optimum length of time for pasture- pounds following cotton dusted with calcium rice rotations in southwest Louisiana. Rotations arsenate. included 1 year native or improved pasture and On lar^d cropped continuously to rice for 49 1 year rice ; 2 years improved pasture and 1 year years, the average annual yields for successive rice ; 3 years improved pasture and 2 years rice ; 5-year periods during the last 30 of the 49 years and 4 years improved pasture and 2 years rice. ranged from 1,102 to 1,440 pounds per acre. Fluc- Fertilization practices varied according to needs tuations in the 5-year average yields apparently shown by soil tests and good agronomic practices. were due to variations in climatic conditions and In this study, length of rotation and pasture treat- were not the result of depletion of the soil fer- ment had little effect on yield of rice. This is be- tility, according to Jenkins and Jones {65). lieved to be the result of using more fertilizer Walker and Sturgis {153) reported in 1946 and the fact that existing soil conditions were that pasture-rice rotation experiments begun in somewhat better in this experiment than in earlier 1938 and conducted on three soil types of the experiments in the area. Improved pastures in prairie rice area of Louisiana showed that a 12- the rotation increased the yield of beef more than month grazing program could be developed. They 500 percent. stated that up to that time the rotation of im- In reporting on an economic appraisal of farm proved pasture with rice definitely was the best practices and rotation programs on Louisiana means found of increasing rice yields and of im- rice farms, Mullins {93) in 1954 pointed out that proving the soil productivity of the area. Also, the most common rotation at that time was 1 year where proper management practices were used, of rice and 1 or 2 years of native pasture. Some yields of beef from improved pastures exceeded longer rotations were being used such as rice 2 92 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE years and improved pasture 3 or 4 years. In the crops the previous year. Where rice followed longer rotations a smaller proportion of the crop- rice, a few operators burned the stubble to facili- land was in rice each year, but the increase in tate land preparation for the second year of rice. rice yields was sufficient to maintain about the In these longer rotations, the major crop grown same total volume of rice production on the in rotation with rice was soybeans. The next farm. most commonly grown crop was wheat. On a A rotation of rice for 2 years and temporary small number of farms, oats were grown. pasture for 3 years gave an average rice yield of Where adequate surface drainage was avail- 2,600 pounds per acre, as compared with 1,950 able, wheat and oats were well adapted to the pounds per acre from rice alternated with native clay soils of the I'ice farms in Mississippi. These pasture in a 2-year rotation. When rice was small grain crops fit reasonably well into the rice grown for 2 years following 3 years of tempo- rotation programs. They usually were seeded on rary pasture, the estimated average rice yield land that had lain idle and had been fallowed the for the 2 years was about 3,250 pounds per acre. previous summer or on land from which early- Leaving land in pasture for a fourth year added maturing soybeans had been harvested. Small an additional 150 pounds per acre to this aver- grains following soybeans did not always pro- age rice yield. In addition, the quality of rice duce satisfactory yields because there was only a improved and this, combined with higher pro- short time to prepare a suitable seedbed. This duction of beef feeding on the temporary pas- sometimes delaye

were obtained on the improved pastures, as com- ^ EVATT, N. S. Texas A. & M. University, Beaumont, Tex. pared with less than 50 pounds on unseeded, un- 1964. [Correspondence.] 488-871 0—73 7 94 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE ping of short-season rice varieties in Texas. They rotation involves rice; spring- or early summer- reported that the technique appears to be a prac- plowed fallow; and fall-sown wheat, oats, or tical means of increasing rice yields, provided oats and vetch. Another includes rice; spring- varieties that mature the first crop in 100 to 105 sown grain sorghum or field beans; and fall- days are grown. It requires about 70 to 80 days sown wheat, oats, or oats and vetch. Safflower to produce the second crop if all operations are may be grown in rotation with rice but produc- properly executed. tion usually is more successful when this crop is Eatooning now appears to be a practical and grown the second year after rice rather than the possibly profitable operation in United States first. areas having a suitable climate, but expert man- Two rotational cash crops are sometimes har- agement is required. Evatt and Beachell (32) have vested the same year. Wheat or oats may be shown that the stubble of the first crop should be followed by irrigated grain sorghum or field at least 16 to 18 inches high. Leaving a shorter beans. For successful double cropping on rice- stubble delays recovery. Significant yield increases land, all operations must be expertly timed. have been obtained from the ratoon crop by ap- Other crops used occasionally on the better plying up to 120 pounds of nitrogen per acre quality, medium-texture riceland soils include immediately after the first harvest. The stubble sugarbeets, melons for seed, tomatoes, and alfalfa. is reflooded when regrowth is 18 to 20 inches high. Williams, Finfrock, and Miller {162) reported Using the early variety Nato, these authors re- that purple vetch (Vicia atropurpúrea Desf.), ported 4,048 pounds per acre from the first crop burclover {Medicago hispida Gaertn.), horse- and 2,382 pounds per acre from the ratoon harvest, beans {Vicia jaba L.), and field peas {Pisum for a total of 6,430 pounds per acre from a single arvense L.) were commonly used in ricelands for seeding. winter-grown cover crop and green manure. They reported that leguminous green manures were California grow^n and turned under on about one-fifth of California's riceland and that the practice is No clear-cut rotation pattern has yet become expanding. established for California riceland (72) for the following reasons : Land Leveling and Seedbed Preparation (1) The soil used for rice production generally is inherently quite fertile. Hence, with the rela- Jones and others {73) reported that most land tively short cropping history as compared with on which rice is grown is comparatively level, that of other rice States, soil nutrient depletion with a gentle slope toward the drainage chan- has not yet proved such a limiting factor as to nels. The cost of developing for irrigation lands make crop rotation necessary. with from 0.01- to 0.50-percent slope usually is (2) No serious rice disease has to date forced economical. A competent surveyor is employed a rigid crop rotation program for disease control. to locate the irrigation canal, drainage ditches, An increasing incidence of stem rot, however, un- and field levees. Improper location of canals^ doubtedly will dictate an adjustment in California ditches, and levees may cause serious losses, be- rice-cropping practices. cause it may result in faulty irrigation and poor (3) The rapidly advancing knowledge of weed drainage. Irrigation canals should be large enough control, crop fertilization, and other improved to supply ample water promptly when needed. cultural practices has made it possible to crop Drainage ditches should likewise be large enough ricelands for a period of 3 to 10 years continu- to dispose of water rapidly. ously and still obtain increasing yields. In 1920, California's average rice yield was 2,295 pounds Land Grading and Leveling per acre; in 1969, it was 5,525 pounds per acre. A number of cash crops are well adapted for A general discussion of land leveling for irriga- growing in rotation with rice in California. The tion has been published by Bamesberger {H). ones most commonly grown are those providing Technical reports of land grading (land forming) the best immediate economic advantage. One for surface irrigation by Gattis, Koch, and McVey RICE IN THE UNITED STATES 95 {Ifl) and Marr {86) discuss the factors to consider farmers were using land planes or similar land- before land grading is undertaken. These include leveling devices to improve the surface condition the determination of crops to be grown, the suit- of their fields but that this operation usually was ability of the land for surface irrigation, the loca- performed on only a part of their land each year. tion and amount of water available, the capabili- Reynolds {113) reported that land leveling with ties of various types of land-grading equipment, land planes or other land-leveling devices is com- methods of calculating for land grading, and ing into greater use in Texas. He emphasized that methods of computation of cut and fill. leveling increases the uniformity of contours be- In developing land for rice, it usually is custom- tween levees and results in better drainage and ary to move enough earth to eliminate hummocks water depth control. These conditions are essen- or ridges and to fill sloughs and hollows. When the tial to obtain and maintain uniform stands of rice. land is graded for basin irrigation, straight and Land planing or floating of riceland usually is parallel levees result (fig. 29). Properly leveled done after heavy disking, field cultivating, or other land drains more quickly in the spring so that means of working the ground. It can be alter- seedbed preparation can be started earlier. Land nated with other operations until the land is leveling also makes it possible to maintain a uni- suitably level. Often land planing is the last opera- form depth of 4 to 8 inches of water within levees, tion before preparing the final seedbed and seeding to control weeds better, and to drain the Avater (fig. 30, A, B, and C). The field then should be in rapidly if necessary during the growing season suitable condition for accurate surveying of the and for harvest. When a field is properly graded, levees. irrigation levees can be made straight and per- In 1962, Faulkner and Miears found that ridges pendicular to the slope of the land. This practice and depressions can be leveled more easily and reduces the amount of land devoted to levees, de- more efl'ectively by flooding the land and working creases tillage and harvest costs, and usually in- the machinery in the water (fig. 30, D). Because creases yield. After a field has been adequately of the plowsole found in most rice soils, equipment graded, it can be maintained by annual land plan- can be supported under flooded conditions. When ing or floating before seeding the rice. water is used as a supplemental moving agent, Wasson and Walker {167) pointed out that greater amounts of soil can be moved more effi- many types of land-moving machines are avail- ciently than when the field is dry leveled with the able and that results from the majority of these same equipment. are good. MuUins {9^) found that Mississippi Faulkner and Miears {35) described two meth- ods commonly used in water leveling riceland. One is to build a levee around small land areas and eliminate all other levees within the field. The other, normally used in large fields, is to remove every second levee and level within the remain- ing areas. With either method, the soil is worked into a loose seedbed and, after being flooded, is moved with a blade and a tractor of 50 horse- power or greater. The blade can be a one-blade land leveler with float removed or it can be any X ,' V other blade approximately 10 feet wide pulled at a right angle to the direction of travel. Faulk- ner and Miears recommended flooding only the land that can be leveled in 1 day because soil be- comes compacted when partly worked and then PN-2985 allowed to remain under water overnight. They FIGURE 29.—Combines harvesting ricefield that has been leveled so that the levees are parallel instead of on suggested leveling the land to within 0.2 foot be- the contour. (Photo courtesy of Agronomy Extension, tween levees for good water control. This allows University of California, Davis.) a minimum water depth of 4 inches and a maxi- 96 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

'é-è'%^'f>^-'^^'i-^ ¿i ■ BN-22023 BN-22040 FIGURE 30.—Leveling or smoothing ricefleUl witli : (A), Large, single-scoop land plane; (ß), large plane with several angle blades ; (O), medium-sized land plane, and (D), small land plane in water. (Photo for A courtesy of Agricul- tural Extension Service, Davis, Calif. Photo for D courtesy of Rice Experiment Station, Crowley, La. )

mum depth of 6.4 inches. They found that water recent years in Arkansas, growers have found it depths of less than 4 inches were ineffective in satisfactory to disk rice stubble immediately after controlling grasses, whereas depths of more than harvest to mix the vegetative matter with the soil. 6.4 inches in Louisiana may injure the rice and Often various adaptations of a large, smooth, result in greater pumping cost for the increased water-filled tank roller rather than a disk harrow volume of water. Leveling in water normally was can be used to break down the stubble so that it difficult where the terrain varied more than 0.6 is in contact with the soil ; then the levees can be foot. No extreme changes in soil characteristics closed and allowed to catch winter rainwater that were observed after leveling in water. Analyses is left on the land until late winter or early spring. showed that the soil generally was more uniform At this time the fields are drained; and as soon in both physical and chemical characteristics after as they are dry enough, the land is plowed with a leveling in the water. One advantage was that moldboard or disk plow and worked with a heavy this uniformity of soil allowed a better determina- disk harrow, or worked with a field cultivator to tion of needed plant nutrients based on analyses of which a spike-tooth harrow is attached, depending soil samples. on the condition of the stubble and the soil (fig. 31, A, B,C, and D). Seedbed Preparation ]Mullins (94) reported that land preparation Jones and others (73) reported that when operations on the heavy soils of the delta in plowed under promptly, rice stubble and weed Älississippi are performed with offset disks, heavy growth decompose readily in the fall and winter tandem disks, and various types of harrows. The if proper drainage is provided. This decomposi- usual sequence is one or two times over with the tion aids in the preparation of a good seedbed. In offset disk, then three or more times over with RICE IN THE UNITED STATES 97 the tendem disk at varying time intervals. Spike- in the fall or early spring. If rice is to follow tooth or spring-tooth liarrows are commonly used soybeans or similar cultivated crops, only a lim- after each cutting with the disk or for the final ited amount of land preparation may be neces- operation before seeding. sary because the soil may be in fairly good physical The seedbed for rice is prepared much the same condition after the cultivated crop has been har- as the seedbed for other small grains. The pro- vested. Reynolds (113) reported that early ex- cedures generally followed are given by Davis periments on time and depth of plowing at the {^6), Finfrock and Miller (37), Jones, Davis, and Beaumont, Tex., station indicated that soil Williams (72), Jones and othe'rs (73), Oelke, plowed 5 to 8 inches deep in the fall or spring Morse, and Mikkelsen (101), and Reynolds {1J3). produced somewhat larger yields of rice than The primary aim is to destroy weeds and to pro- soil plowed only 2 inches deep. There was little vide a suitable seedbed. Whether it should be diffei'ence in yield of rice between summer, fall, rough or mellow on the surface will be governed and spring plowing. However, summer, fall, or by the seeding method used. If the rice is to be early winter plowing had an advantage over sown in the water, the soil surface before flooding spring plowing in that labor was distributed over should be dry and rough; if rice is to be drilled, longer periods. Laud tliat was plowed in the fall the soil surface should be mellow and smooth. formed a better tilth and was easier to prepare Depending on the crop that precedes rice in for seeding by disking and liarrowing the follow- the rotation, it may be desirable to plow the land ing spring than land that was plowed in the

,.eJf ^ -Í3Í '*' ¿■fit * ^íj[¿H»

f. BN-22024 FIGURE 31.—Preparing the seedbed: (A), Water-filled tank roller used to force rice stubble into the soil; (B), mold- board plow; (C) field cultivator with spike-toothed harrow attached; and (D), disk harrow. (Photo for D courtesy of Agronomy Extension, University of Oalifornia, Davis. 98 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE, spring. Depending on the condition of the soil, farmers may find that disposing of crop residue by riceland generally is plowed with heavy disk or burning will be legislatively restricted or prohib- moldboard plows (fig. 31). ited. Eesearch findings of Williams and others Disk plows generally are used when the soil {163) suggest an alternate method of disposal. In is dry and hard. When the soil is moist and in a 5-year test at one location, the incorporating of good condition for plowing, moldboard plows all straw into the soil did not decrease grain yields are very satisfactory. In Arkansas, the use of although the straw^ ranged widely in nitrogen con- moldboard plows has decreased in favor of vari- tent. The net nitrogen immobilization was 0.54 per- ous types of disk plows, which can cover the cent of the original dry weight of the straw as ground more rapidly and which may leave a determined by grain yield responses over single smoother surface that will require less leveling or growing seasons. When applied on nitrogen-defi- floating. cient soil, straw with higher nitrogen content in- Keynolds {113) reported that summer plowing creased grain yield but straw with low nitrogen frequently is practical in Texas on fields infested content decreased grain yield. Supplemental ni- with red rice or other weeds. After summer trogen applied in the form of either urea or vetch plowing, the land may be leveled and disked or gave increased grain yields in the presence of both harrowed as needed for weed control. Land high- and low-nitrogen straw. No significant im- plowed in the fall usually is left rough until mobilization of supplemental nitrogen resulted spring, then it is disked and harrowed prepara- from straw application. tory to seeding. It is desirable to open drainage A better seedbed can be made on spring-plowed furrows through a field after the last tillage soils if they are allowed to dry for 7 to 10 days operation in the fall to provide adequate surface after plowing before the final seedbed is prepared. drainage. Some farmers wait until the first Davis {26) reported that water-seeded rice germi- heavy rain of the fall when some water is stand- nates better, has more seedling vigor, and produces ing on the land to make the drainage furrows a lieavier crop when sown on a seedbed that is dry so that natural drainage channels can be located. during the finishing operations. A seedbed pre- A small grader, a bedder, or a one-bottom plow pared under moist conditions for water seeding may be used to make this drainage furrow. Usu- induces algae (scum) development, increases weed ally it is not necessary to plow in the spring the problems, and frequently results in a poor stand fields that were plowed the previous summer or because of poor germination and seedling vigor. fall, except on poorly drained soil or during Finfrock and Miller {37) reported that a good seasons of heavy rainfall. winter-grown vetch cover crop aids in drying out In general, land plowed in the spring should riceland soil, thereby making it possible to pre- be disked and harrowed as soon as possible after pare the seedbed earlier. Cover crops turned un- plowing to break up any large lumps and clods, der for green manure are usually plowed down to prevent baking or crusting, and to avoid sub- with a moldboard or a diskplow. For best results sequent difficulty in preparing the seedbed. In in California, the plant material should be com- Texas, experience has shown that heavy soils, pletely covered with 4 to 6 inches of soil, accord- such as Beaumont clay and Lake Charles clay, ing to Williams and others {161^ 162). generally require more subsequent tillage, such as disking and harrowing, to obtain a desirable Seedbed Preparation as Related to Method seedbed when plowed in the spring than when of Seeding plowed in the fall or early winter {113). Eice may be seeded either by drilling on dry Usually California ricelands are spring plowed ground, or by broadcast seeding with an endgate to a depth of 4 to 6 inches, after the stubble of seeder or airplane on dry ground or in flooded the previous crop has been partly reduced by fields. The final seedbed preparation is influ- burning. The straw may be burned in the fall enced by the method of seeding to be used. If immediately after rice harvest if dry weather fields are to be water seeded, growers sometimes continues long enough. are able to prepare a fairly good seedbed in the Because of strong public interest in "clean air," late fall or winter and to erect their levees at RICE IN THE UNITED STATEiS 99 this time. When this is possible, very little land levees are considerably higher and larger and are preparation other than harrowing is necessary in not seeded. the spring. In the Southern States, if the field Some farmers have been successful in preparing is to be dry seeded with a drill or with an end- their land in relatively stale water by drawing gate seeder or broadcast by airplane, the final spring-tooth or spike-tooth harrows through the preparation of the field includes working over fields in the mud or in fairly deep water. Other the levees w^hile the field is worked. When rice growers who have attemped this method have is to be drill-seeded, the final preparation just found it very unsatisfactory. ahead of the drill usually is done with a spring- tooth or disk harrow behind which is drawn a Construction of Levees spike-tooth harrow. This gives a mellow firm seed- Eicefields are divided by levees into subfields, bed, and the moisture is held near the surface so called paddies or bays (cuts). Levee construction that the seed usually will germinate soon after is an important operation in preparation for grow- seeding without irrigating the field. A roller- ing rice because levees are the key device for regu- packer may be used in order to break up clods lating water depth in ricefields. Levees must be before drilling and to firm the soil after drilling located accurately and must be well constructed in to help retain moisture. Usually, the levees are order to maintain a uniform depth of water within reseeded after they have been partly rebuilt, and each paddy (fig. 32). the seed then is covered by the final building of The levees are constructed on the contour, that the levees with a levee disk or a pusher-type levee is, on lines of equal elevation. They should be maker. located by an experienced surveyor or operator If rice is to be broadcast seeded on dry ground, who uses an accurate instrument. Because a smooth the final seedbed preparation leaves the surface soil surface is needed for accurate location of the rough and somewhat cloddy. The seed usually is levees, the surveying should be done immediately covered by a shallow working with a spring- after the field has been floated. On flat land, the tooth, spike-tooth, or disk harrow. Seed on the difference in elevation between levees is 0.1 to 0.2 levees is covered by the final working with the foot, and on steep, sloping land it is 0.3 foot {38^ levee disk or pusher. In most cases, precipitation 73), In fields where the levees run parallel to the following seeding is necessary to bring about ger- direction of the prevailing wind, it is desirable, mination and emergence of the rice seedlings. especially if the paddy is large, to build wind If a field is to be water seeded, the final seedbed levees at right angles to the direction of the pre- preparation depends somewhat on soil type. On vailing wind. These levees are not tied into the sandy or silt loam soil, a mellow, firm seedbed simi- border or contour levees. Their function is to re- lar to that for drilling should be prepared. Levees duce the effect of the wind and thereby decrease are constructed and are seeded just before the final wave action and reduce the possibility of levees working with a levee disk or pusher. The remain- being washed out. der of the field is worked between the levees with a The levee should be compact and high enough to spring-tooth harrow, which leaves fairly deep fur- hold the water at an average depth of 3 to 6 inches rows into which the seeds fall as they settle in the paddy. In the Southern States, the levees through the water. have gently sloping sides, so that they may be When rice is water seeded on clay or very fine crossed w^ith cultivating and harvesting equipment. silt loam soil, the seedbed should be fairly rough Here, the levees are completed after the rice is with a clod size ranging up to 4 inches in diameter. drilled. Levees of this type are considered very This is done in California by harrowing twice with efficient, for an entire field can be cultivated, a heavy spike-tooth harrow followed by dragging seeded, and harvested as a single unit. with a heavy wooden drag. A rou^h seedbed helps In California, the levees are higher and have prevent drifting of water-sown seed. As the clods steep sides so as to hold a deeper flood; and the slake down after flooding and seeding, a fine film areas between pairs of levees must be harvested as of soil may cover the seed. In the South, the levees a field or unit. This increases farming costs because usually are seeded separately. In California, the it increases machinery-maneuvering time. In addi- 100 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE,

tion, the land area devoted to high levees is non- productive. In the Southern States, the base of the levee commonly is made by plowing one round with a 3-bottom plow. Levees are constructed with a levee disk usually mounted on the rear of a large tractor, or commonly with a single- or double-blade pusher on clay soils. The levee base is built as early as possible, so that it will become compact before flooding time. This is especially important on heavy clay soils on which levee construction often is difficult. Compact, well-settled levees reduce seepage and are less likely to be washed out by excess water from heavy rains. Well-constructed levees, properly repaired after seeding, facilitate irrigation and eliminate much expensive hand shoveling. Pickup blades or scoops mounted on small tractors are used to close gaps at the ends of levees and to help install levee gates. This elim- inates much of the hand labor formerly required. In California, a V-type diker or soil crowder that is 14 to 16 feet wide in front and 4 feet wide in the back is used to build levees. Two or more heavy-duty, crawler-type tractors are used to pull the diker. The levees usually are from 30 to 36 inches high when freshly made and settle to be-' tween 16 to 20 inches. Levees usually are made in the fall with a base of 5 to 6 feet and are allowed .^^•.^Aái-'r; to settle during the winter. Most of the work of closing the gaps at the water control boxes and ends of the levees is done with a bulldozer or a tractor with a front-end scoop. Scott and others (Wß) reported on the possi- bility of using rice levees made of plastic film instead of levees made of soil. Their studies show that plastic levees are physically and economically feasible. Although the original developmental con- cept was to di'ape plastic sheeting over a line of contoured stakes in the ricefield, the system has evolved commercially into low earthen levees cov- ered by plastic sheeting. In 1970, privately devel- oped machines under commercial contract were used to construct plastic-covered earthen levees in California at a price competitive with the cost of constructing large soil levees. However, presently BN-22035 available plastic sheeting is essentially nonbio- FIGURE 32.—Making levees with (A) dislv, (B) puslier, degradable and tends to accumulate in the soil and (C) plastic. (Photo for A courtesy of Agricultural Extension Service, Stuttgart, Ark. ; photo for C courtesy as nondecomposing trash. This problem remains of Agricultural Extension Service, Davis, Calif.) to be solved. RICE IN THE UNITED STATE

Seed and Seeding poorly may result in uneven stands, uneven ripen- ing, and low field yields of poor milling quality, The choice of seed and the use of suitable seed- Oelke and others {100) studied the effect of ing methods are an important part of rice culture. moisture content at harvest on grain yield, germi- To achieve a high level of production, the variety nation percentage, and seedling vigor. They found to be grown must be adapted for the area. After that grain yield increased slowly as moisture con- deciding on the variety, it is necessary to select a tent decreased from 31 to 19 percent but that ger- lot of rice seed that is free from varietal mixtures, mination percentage increased steadily. Seedling does not contain red rice and weed seed, is high in vigor and speed of emergence through 10 centi- percentage of viable seed, and has high bushel meters of water was faster when grain moisture at weight. All rice-producing States now have a harvest w^as below 20 percent. Seedling growth in rice seed certification program designed to pro- a germinator was earlier and faster as grain mois- vide high-quality seed for the rice industry. ture decreased to 13 percent. Before 1941, when combine harvesting, artificial Smith {135) reported in 1940 that seed lots drying, and bulk storage were begun in the south- used on 29 farms were examined, and only four ern rice area, there was little incentive to develop were found to be free of weed seeds and red rice. a seed rice industry. Varietal mixtures resulting Today, most rice seed is thoroughly cleaned and from handling were not serious when binders, graded. Consequently, it is free of weed seed and stationary threshers, and sack storage were used. has only a trace of hulled or poorly filled grains. Seed rice usually was saved from fields or parts Much emphasis has been placed on the impor- of fields of known varietal purity. Weed seeds tance of eliminating red rice from planting seed and trash were removed by fanning mills and (^, 7^). Eed rice is objectionable because the red disk or cylinder graders usually owned and oper- bran is not completely removed in milling. This ated by private or grower organizations. How- results in an unattractive appearance when milled. ever, as combine harvesting, artificial drying, and Also, the grain size, shape, and milling quality are bulk storage developed, varietal mixtures became inferior. Eed rice tillers profusely and shatters serious. It was at this time that seed production easily, and seeds that have become buried in the and processing developed into a specialized busi- soil have been known to remain viable for several ness. Today, much of the seed rice sown in the years {19^ 4-4-)* When rice seed containing red rice United States is given specialized attention during was seeded several years in succession on the same growing, harvesting, drying, and processing. As a land and red rice was not controlled, over 50 per- result, the seed rice industry is well developed. cent of the harvested grain was red rice (^). In a Foreign demand for seed of high quality has fur- ricefield in Louisiana, over 500 red rice grains were ther stimulated the development of the seed rice found in the top 6 inches of each square foot of industry. Substantial quantities of seed rice are soil.^ exported annually to Central and South American The importance of using seed free from red rice countries. cannot be overemphasized because as long as red rice is being sown, it cannot be eliminated from Seed Quality the fields. Based on a seeding rate of 80 pounds per acre, seed containing 5 red rice grains per pound Numerous workers have stressed the desirability would probably result in approximately 200 red of using seed rice of good quality {18^ 27^ 7^, 135). rice plants per acre. At the Beaumont Eice-Pasture Good seed should be well matured and free from Eesearch and Extension Center, land was cropped red rice, immature and hulled or broken grains, 2 consecutive years, in 1946 and 1947, and seed was seed of other varieties, and weed seeds. The seed used that contained approximately two red rice should be cleaned and graded to remove hulls, grains per pound. The grain harvested the first trash, and other foreign matter. Also, good seed should germinate satisfactorily and should pro- ^ BAKER, J. B. Louisiana State Univ., Baton Rouge, La. duce strong sprouts. Using seed that germinates 1963. [Correspondence.] 102 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF A(3}RICULTURE

year contained about 18 red rice grains per pound WARNING and that harvested the second year contained over 125 grains.^ The materials recommended here for treating rice seed should be considered poisonous to man and animals. Care should be taken Seed rice free from red rice was nearly non- in handling and using them. Read the label placed on each con- existent until 20 years ago. Screen graders, which tainer by the manufacturer and follow his instructions regarding effectively grade seed rice on a grain diameter safety measures. All workmen operating seed-treating equipment basis, have been effective in removing the broader should be carefully taughl- how to use the chemicals and should be red rice grains from long-grain varieties. The warned against carelessness. Sacks of treated seed should always be properly labeled. Care should be taken to prevent any treated screen graders, along with the production and in- seed from being used' as food or feed. crease of seed stocks free from red rice, have ef- fectively controlled red rice. In a few instances, slender red rice grains that cannot be separated and seedlings, according to Webster and others from long-grain varieties have been observed. {158), Although prevalent throughout the rice- Whenever a seed lot that contains such types is producing areas of California, damage generally observed, it should be discarded immediately. is more severe when temperatures are cool and un- Ked rice grains are difficult to remove from favorable for the growth of rice. Achlya klebsiana short- and medium-grain varieties because of the or Pythium species were usually isolated from de- similarity in diameter between those varieties and cayed seeds and infected seedlings. Outgrowths of red rice. Since red rice grains cannot be effectively whitish hyphae characteristically radiated from removed from the medium- and short-grain vari- cracks in the glumes or from the collar of the eties, it is essential to use seed free from red rice. plumule. In these tests by Webster and others, the The origin of high-quality seed rice and the cer- standard California varieties (Caloro, Calrose, tification of rice seed are discussed in the section and Colusa) were equally susceptible. Optimum on "Rice Breeding and Testing Methods in the temperature for growth of A, klehsiana isolates in United States," p. 22. culture was from 27° to 30° C, but pathogenicity did not differ significantly between 20° and 30°. Source of Seed A mean of 52.8 percent of the plants became Jones and others {71) studied the effect of en- diseased over this range. Pythium isolates grew vironment and source of seed on yield and other well in culture at 30° but were significantly less characters of a group of varieties at four rice ex- pathogenic (17.2 percent of the plants infected) periment stations in Arkansas, Louisiana, Texas, than at 25° (48.6 percent of the plants infected) or and California. They concluded that seed source at 20° (95.5 percent of the plants infected). These had no appreciable effect on grain test weight, ger- results are consistent with field observations. Cool mination of seed, average grain and kernel weights temperatures apparently increase disease severity within a variety, proportion of hulls, and milling primarily by an adverse effect on growth of rice quality. In summary, they stated that "Local seed seedlings. of good quality, free of mixtures and weed seed, Mikkelsen and Glazewski {88) reported the is as productive as that obtained from other rice- presence of eight compounds in the hulls of Caloro producing States." rice that diffuse into the embryo and inhibit germi- nation when present in large amounts. Compounds Seed Treatment identified included sinapic acid, vanillic acid, Whether drilled into a moist seedbed or broad- p-hydroxybenzoic acid, ferulic acid, p-coumaric cast in water, rice seed after sowing is soon subject acid, Índole 3 acetic acid, p-hydroxybenzalde- to attack by organisms widely dispersed in the soil hyde, and vanillin. In low concentrations^ these and water. chemical substances stimulate germination and en- Eice seeds broadcast onto a flooded field often suing growth. Leaching the seed by soaking in are quickly infected by diseases that damage seeds water or in a water-sodium hypochlorite solution reduces the concentration to stimulatory levels.

*'BEACHELL, H. M. Rice-Pasture Research and Exten- Garrison {Jfi) reported on work with other sion Center, Beaumont, Tex. 1963. [UnpubUshed data.] small grains conducted by Earhart. Results RICE IN THE UNITED STATES 103 showed that there is a definite need for all small in June, there was a risk that the crop might fail grain seed to be completely processed, including to mature if there was an early frost. However, thorough cleaning and treating. For every 100 with the development of the very early-maturing field-run seeds put in the ground, only 58 healthy varieties Belle Patna and Vegold, mid-June seed- plants were produced. When seed from the same ings appear to be relatively safe, even in northern lot was cleaned, 65 healthy plants were produced ; Arkansas. Johnston and others {67) suggested and when the seed was both cleaned and treated, that these two varieties be seeded in northern 72 healthy plants were produced. Field experi- Arkansas from about June 1 to 10, in central ments comprising two rice varieties, sown at two Arkansas from about June 1 to 20, and in southern dates at each of three locations for 2 years, showed Arkansas from June 1 to 25. Johnston, Cralley, an average survival of 61 percent for seed treated and Henry {69) emphasized that a relatively with two commonly used seed-treatment chemicals, "safe" seeding date may depend to a considerable compared with 52 percent for untreated seed {12). extent on anticipated water management and fer- tilizer practices. They pointed out that heavy Time of Seeding rates of nitrogen fertilization, particularly when applied late, tend to delay maturity. Seeding of rice in the United States begins Results from date of seeding experiments indi- when the weather becomes warm enough for germi- cate that there is a comparatively long period in nation and seedling growth. In Arkansas, Missis- the South during which rice can be sown and still sippi, Missouri, and California, where the produce satisfactory yields. It is possible and growing season is comparatively short, seeding often advisable to spread the seeding time of cer- usually is done in April and May, within a period tain varieties so that the harvest can be extended of 3 to 5 weeks. Near the Gulf Coast in Louisiana over a longer period. However, Adair (^), who and Texas, rice can be sown from early March to studied the effect of time of seeding on yield and late June—although most seeding is done in April. milling quality in Arkansas, concluded that most The length of seeding period depends mostly of the early and midseason varieties produced rice on the length of life cycle as influenced by the of better milling quality when they matured late photoperiod and heat response of the available in September or early in October than when they varieties. Essentially it is the difference between matured before September 15. Jodon {66) stated the length of the growing season for rice in an area that since the approximate number of days re- and the shortest period of time required to ma- quired from seeding to maturity is known for all ture a crop. commercial varieties, it may be desirable to base Seeding time also is influenced, directly or in- the date of seeding on the number of days re- directly, by favorable weather conditions for land quired for maturity, so that the rice will mature preparation and seeding operations, methods of in the fall when the weather is usually dry and seeding, fertilizer practices, availability of fresh pleasant for harvest. Bice maturing after the water, temperature of water, cold tolerance of temperatures have lowered somewhat may be of varieties, and time of maturity of varieties in re- better milling quality. When rice matures during lation to date of seeding. extremely hot weather, considerable sun checking Adair and Cralley (5) stressed that the proper may result and low head rice yields may be pro- seeding date for each variety is important. They duced when the grain is milled. stated that no rice variety should be sown until Chambliss {18) stated that in the prairie areas the mean daily temperature rises to about 70° F. of Louisiana and Texas, most of the rice is sown They and Johnston, Cralley, and Henry {68^ 69) from April 1 to May 15. However, he pointed reported that because of the relatively long grow- out that May 1 was approximately the best date ing period of several of the varieties then avail- able, it was necessary to seed these varieties in for sowing rice on the prairie because of the cold Arkansas by late April or early May. Other, weather that sometimes prevails during April. shorter season varieties had a wider range of seed- According to Jones and others {73)^ rice usually ing dates. However, when any variety then avail- is sown in the Southern States from April 1 to able was sown in Arkansas after the first week May 30. But if conditions are favorable, rice is 104 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE!

sometimes sown in March and as late as June 30. gestion at the driers. Also, early seeding of short- Rice germinates more quickly when sown in the season varieties usually allows time for the produc- late spring, when temperatures are relatively tion of a stubble crop in southern Louisiana and high, than when sown earlier. Also, late sowing Texas. Because of the somewhat shorter growing has an advantage in that the weeds that have season and cooler fall temperatures, results and started growth can be killed by cultivation before observations indicate that such double cropping the rice crop is sown. (ratoon cropping) is not practical in Arkansas Reynolds {US) stated that time of seeding rice with varieties presently available {67). Optimum in Texas ranges from March 1 to late June but seeding time in Arkansas is late April and early that most of the rice acreage is seeded in April May in most years and for most varieties. and May. He pointed out that the actual time However, even the customary April seeding of seeding may depend on several factors, such often is subjected to unfavorable conditions for as the weather, method of seeding, soil condition, establishing stands. Varieties that have short grow- and maturity group of the variety. Some varie- ing seasons mature during hot weather if seeded ties can tolerate cool weather in the spring better at the average time, and in some seasons the quality than others and will thereby produce better stands is poor. Localized summer windstorms and rain- when seeded relatively early. He further re- storms may cause lodging »with resulting loss of ported that the yield of certain varieties showed yield and increased cost of combining. Often at the a marked decrease when seeding was delayed peak of the season more rice is ready for harvest beyond certain dates. He concluded from tests than the driers have capacity to process, and the at Beaumont that the late-maturing varieties rice that waits in the field is in danger of should be seeded as early as practicable to insure deteriorating. satisfactory yields. Rice matures in a shorter time when sown late In California, the period for seeding rice is more than when sown early. Delaying seeding or using concentrated than it is in the South. Some rice is varieties that require long growing seasons per- sown in California soon after April 1 and some as mits harvesting to be done in the normally cool late as June 15, but most of it is sown from April 15 autumn weather when there may be fewer showers to May 15 {37). When rice is sown as late as May 30 and when the driers are less likely to be crowded. in California, it should be fertilized at a moderate Later maturing rice usually is better quality as rate, and preferably a short-season variety such as long as the temperatures are not too low for nor- Colusa should be grown. mal development of the grain. Delaying seeding Very early seeding entails certain hazards such also allows time for additional cultivation to re- as loss of stands due to seedling blight or drown- duce weed and red rice infestation. On land that ing out of drilled rice. Early growth is slow, pro- has been in winter or perennial pasture crops, a de- longing the period from seeding to harvesting. lay in date of seeding provides a longer grazing Isolated early-maturing fields may be attacked by period and makes possible the preparation of a concentrations of insects, rodents, or birds. Eice better seedbed. that matures early during hot and dry »weather Late seeding also has disadvantages. There may tends to have lower milling quality. be a greater possibility of injury by root maggots, Relatively early seeding in the Southern States armyworms, and stem borers. Weather conditions helps assure an adequate supply of water to mature may be more favorable to the blast fungus at the the crop before a summertime shortage of water or, time the young plants are susceptible. Stem rot in limited areas where it occurs, before the intru- may be more severe late in the season. The crop is sion of salt water. Water is available for early- exposed to early fall storms, likely to be the most seeded rice in all areas, although in California damaging of the season. Cold weather may reduce the water may be too cold. Early-seeded rice quality and yield if harvest is extremely late. Iso- may tiller more, may compete more effectively with lated late-maturing fields probably are more ex- weeds, and may escape insects and diseases to some posed to damage by concentrations of birds and extent. When rice can be harvested early in the other pests than are extra-early fields. However, season, it usually is possible to avoid seasonal con- bird damage may be severe in isolated early fields RICE IN THE UNITED STATES 105 if local populations of blackbirds or sparrows are rates of seeding, higher yields resulted from the present. lower rates of seeding. The highest average Clearly, there is no ideal time for seeding. In yields were obtained from sowing approximately Louisiana and Texas especially, where the seeding 70 pounds of seed per acre. period is long and there are several varieties to Simmons {WJi) reported that later studies at choose from, the date of seeding demands careful the Arkansas Eice Branch Station at Stuttgart consideration. This is true whether one field is to be indicated that the best rate for drill seeding was seeded to a single variety or a large acreage is to be 90 to 110 pounds per acre and that about 110 to alotted among two or more varieties. Where large 135 pounds per acre was best when the seed was acreages are seeded to one variety, seeding dates broadcast. should be spread out in order to prevent congestion Chambliss {18)^ reporting on early experiments at driers at harvest. at the Crowley, La., Eice Experiment Station, found that the largest yields and the best quality Rate of Seeding of milled rice were obtained from the Honduras Factors that enter into a determination of the variety by drilling 80 pounds of seed to the acre. proper rate of seeding include seed size and qual- Other varieties available at that time could be ity, fungicide treatment, condition of the seedbed, seeded at a slightly lower rate. He pointed out fertility of the soil, date of seeding, and variety. that less seed may be used when the crop is sown Seed treatment with an approved fungicide pro- in late May if the seedbed is well prepared be- tects germinating rice seed and thus makes it pos- cause better germination is obtained with higher sible to use less seed per acre. In the southern rice temperatures prevailing. The quantity to be sown area, the seeding rate is about 90 to 110 pounds depends on the method of seeding, variety, char- per acre when drilled and about 115 to 150 pounds acter of the seedbed, soil fertility, and vitality or per acre when sown broadcast on dry soil or in the germination of the seed. If rice is broadcast on water. In California, when rice is sown in the wet land or on a poorly prepared seedbed, the rate water, seeding rates average about 150 pounds and of seeding should be increased. If seed of low range from 125 to 200 pounds (dry-weight basis) vitality is used, then the seeding rate should be per acre. increased accordingly. Seeding at too low a rate The number of rice seeds per pound ranges from resulted in excessive tillering, irregular ripening, 14,000 to 22,000, depending on the variety. With and reduced grain yields. a seeding rate of 150 pounds per acre, seeds are Eeynolds {US) stated that in Texas the rate of sown at the rate of about 50 per square foot. Excel- seeding ranges from 60 to 125 pounds per acre, lent yields have been obtained from populations with the average about 90 pounds. The rate of ranging from 8 to 30 plants per square foot. Seed- seeding varies greatly in different parts of the ing rates that provide plant populations between rice belt of Texas. In the more humid areas, these extremes apparently do not influence yields. higher rates of seeding are used than in the west- Extremely dense stands lodge more readily than ern counties where rates as low as 60 pounds per do optimum stands. Eice in dense stands heads and acre frequently are used. At Beaumont, from 1914 matures more uniformly than does rice in thin to 1918, seeding 100 pounds per acre produced stands with abundant tillering. Weed control is slightly higher yields of rough rice than seeding more difficult in thin stands, so seeding rates should 60 and 80 pounds per acre. Experiments were con- be higher in fields infested with weed seed. Rela- ducted there from 1950 to 1952 to determine the tively high seeding rates usually are used on land optimum rate of seeding rice under several levels on which many rice crops have been grown of soil fertility. Bluebonnet 50 was broadcast at previously. rates of 45, 90, 135, and 180 pounds per acre. In early tests in Arkansas, Nelson {98) tried Fertilizers were applied on the surface with a 11 rates of seeding, using recleaned seed, seeded fertilizer-grain drill at the time of seeding. There with a grain drill. The results showed no marked were no significant differences in the average preference for any rate. Although denser stands yields of rough rice from the seeding rates of and fewer weeds were obtained from the higher 90, 135, and 180 pounds per acre, but the yields 106 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE from these rates were significantly higher than they insure even maturity. Grain maturing those from the 45-pounds per acre rate. As in evenly will have a better grade, and consequently earlier experiments, the optimum rate of seeding will usually command a better price. He further was about 90 pounds per acre. Where weeds are stated that when rice was seeded with a grain troublesome, heavier rates of seeding usually give drill, a depth of 1 to 2 inches normally should be better results. used. Deeper seeding sometimes resulted in a Jones and others {73. 7Ji) summarizing rate of poor stand, especially if the soil was clay tex- seeding experiments with drilled rice under tured, cold, or inclined to crust after rains. favorable conditions, indicated that 80 pounds of eTones and others {73) also reported that shal- seed per acre usually is sufficient to give good low seeding is preferable to deep seeding on a stands. Yields are seldom materially increased rough seedbed or a fine mellow seedbed that is and sometimes may be reduced if more than 80 inclined to crust after rain. Surface crusts may pounds per acre is sown. They pointed out that be broken by a light harrowing or by irrigating under ordinary conditions, 90 to 100 pounds of to permit the seedlings to emerge. recleaned viable seed sown with a drill or 110 to Nelson {99) reported in 1944 in Arkansas that 150 pounds sown broadcast is sufficient to give seeding rice broadcast on the surface of the soil good stands. They stated that the rate of seeding and irrigating immediately gave large increases should be sufficiently high to produce stands that in yields of rice over the usual method being used are thick enough to help check weed growth and at that time. He pointed out that even larger also to prevent late tillering. The latter often re- average increases in yield were obtained when the sults in irregular ripening and grain of inferior soil was flooded and the seed broadcast in the quality. water. He indicated that both methods could be used to produce higher yields of rice if the prac- Method of Seeding tical difficulties involved in seeding large areas by these methods could be overcome. The yield Several methods are used to seed rice in the increase apparently was due to the control of United States. On dry soil rice may be sown li/^ grass. to 2 inches deep with a grain drill, or broadcast Mullins {9Ji) found that in Mississippi, where and disked or harrowed to cover. When soil ground equipment w^as common, rice was seeded moisture is not sufficient for germination and after the first levee-building operation was per- growth, the field may be flushed. The water is formed and before the levees w^ere raised to the drained immediately because the rice seedling desired height. Seeding was done with endgate cannot emerge through both 1 to 3 inches of soil seeders by approximately 50 percent of the farm- and 4 to 8 inches of water. Eice also is broadcast ers who were surveyed, and grain drills were used in water by airplanes (fig. 33). Several modifica- by the remainder of those using ground equip- tions of the water-seeding method are used. Rice ment. By 1958, about 20 percent of the farmers is not transplanted in the United States. seeded some rice in the water with planes, and Simmons {12Ji) reported in 1940 that the two the practice appeared to be more common in 1959. methods of seeding being used in Arkansas at the Water seeding often was used during the later time of his report were the broadcast and the drill weeks of the planting season when quick germi- methods. The broadcast method had been most nation and effective grass control are particularly popular because it did not require much machin- important. Some farmers reseeded their levees ery, and it was used extensively in new rice areas. after the final levee-building operation, but the He pointed out that drilling was being used widely majority did not. on old riceland. Advantages of drilling over Reynolds {113) reported that a depth of 1 to 2 broadcast seeding were that less seed was required inches was best when rice was seeded with a grain for a good stand and that the seed could be put drill. In experiments conducted at Beaumont, down at a uniform depth and rates, which make a Tex., from 1914 to 1918, placing the seed 2 inches more uniform stand possible. Simmons pointed deep produced slightly higher yields of rice than out that uniform stands are important because placing the seed 1 or 3 inches deep. There RICE IN THE UNITED STATES 107

BN-2031 PN-2990 FIGURE 33—(A), Airplane being loaded with seed rice, with tanks used for soaking seed In the foreground and rlceflelds ready to be seeded on each side of landing strip; and (B), airplane rotary attachment for distributing seed or fertilizer (Photo for B courtesy of Agronomy Extension, University of California, Davis.)

usually was less rotting from shallow seeding than Calif., in 1929, for reseeding a field of rice. A from deep seeding, especially if irrigation was fair stand of rice and a satisfactory yield were necessary to germinate the rice seed. Seed could obtained. Several growers in California seeded be planted deeper on sandy soils, such as Katy their rice from an airplane in 1930. Now air- and Hockley soils, than on clay soils, such as plane seeding is the common practice on Cali- Beaumont clay and Lake Charles clay. fornia rice farms. Reynolds (113) pointed out that rice usually was The usual practice in California is to soak the seeded in Texas with a grain drill but that aerial seed for 18 to 24 hours, drain for 24 to 48 hours, seeding was increasing rapidly and that 110,000 and seed by airplane into fields that have just been acres were seeded by plane in 1953. He stated flooded to a depth of 3 to 6 inches. Seed rice will that broadcast seeders were used to some extent. absorb tlie maximum amount of water in 18 to 24 On rough, dry seedbeds, rice frequently was hours. Fungicide seed protectant is applied as a broadcast with endgate seeders or with a drill slurry before soaking. with the disks removed. In either case, the land Soaked seed is used instead of dry seed. Soak- was then harrowed after seeding and was irri- ing starts the germination process before seeding. gated. "VVlien seeding was done on dry land, fields More important, because it is heavy, the soaked usually were harrowed after seeding and then seed sinks into place when it hits the water. Thus flushed to bring about germination. The fields were the seed does not drift and luiseeded areas are not resubmerged in time to control the growth of weeds left in the field. Fields remain continuously and grass. flooded until drainage for harvest. The seeding of rice in water was started in Modifications of water-seeding methods broadly California as a way to control barnyardgrass adapted to use in the southern rice area have {Echinocliloa spp.) {6). At first rice was sown been developed. Slusher {130^ 131) extensively with a tractor-drawn endgate broadcast seeder, discussed the use of aircraft to seed rice in but this method was not very satisfactory. Air- Arkansas. plane seeding was attempted first near Merced, Adair and Engler (6) described the water- 108 AGRICULTURE HANDBOOK NO. 289, U.&. DEPT. OF AGRICULTURE seeding method as used in Arkansas as follows: after the 4- to 6-inch depth of water is obtained. The land usually is plowed in winter and a seed- In the test conducîted, there was no significant bed is prepared in the spring by disking two or difference in plant stand densities obtained due three times and then harrowing. Frequently, the to seeding in clear or in muddy water that cleared soil also is tilled to a depth of about 8 inches with within 24 hours after seeding. Likewise, presoak- a field cultivator to provide aeration and space ing the seed showed no significant difference from for applying cool irrigation water that contains seeding with dry seed. It was found that seed dissolved oxygen. Oxygen is essential for early treated with a fungicide or insecticide, or both, had root development. The levees are completed after less tendency to float than did untreated dry seed. the soil is worked wdth the field cultivator and (4) After seeding, the full flood of 4 to 6 the field is then cultivated between the levees with inches of water should be maintained for 5 to 6 a spring-tooth harrow^, which leaves shallow fur- weeks. Lowering the water depth or draining the rows and ridges that help to reduce drifting of field soon after seeding allows grass to become the seed. The levees may or may not be seeded established with the rice, thus defeating the pur- before final going over with the levee disk. pose of water seeding. A rate of 135 pounds of seed Floodgates are then put in, the field is submerged per acre gives a sufficient number of plants, so that to a depth of 4 to 6 inches, and the rice is sown tlie loss due to wave action is much less than loss from an airplane. Seeding is done as promptly as from weeds due to draining the field. possible, because poor stands usually are obtained (5) After a period of 6 weeks or more, the field when the water has been on the field longer than may be drained for midseason nitrogen fertiliza- about 4 days before seeding. tion. The airplane operator is guided by flagmen, A few growers in Arkansas pregerminate the one at each end of the field, who pace off the dis- seed by soaking it before seeding. In one method tance (about 30 feet) that the plane can sow in one tlie seed is placed in a grain cart such as that trip across the field. Depending on the length of used to haul the rice from the combine at harvest- growing season of the variety being grown and the time. The grain cart is filled with water, or water desires of the operator, the water may be drained is kept running through the seed for several hours. from the field after 5 weeks or more to control the Then the water is drained and the rice is left in the rice water weevil, to prevent straighthead, or to grain cart overnight. The following day it is provide dry soil for topdressing with fertilizer (6). angered into the hopper of the plane for seeding. Occasionally, draining at an earlier date may be In another method the seed is soaked in bags in a necessary. However, unless specific conditions exist canal for a somewhat longer period, up to 36 that require early draining, it is advisable to drain hours. late because high grain yields have been obtained when flood water w^as left on fields until shortly Faulkner (S^) compared different practices for before harvest (98), water seeding rice in Louisiana. These included From tests with water-seeded rice on clay soil (1) pregerminated seed compared with dry seed, in Arkansas, Hall (51) suggested the following: (2) cloddy seedbed compared with seedbed that (1) A disk harrow^ is effective in preparing the had been smoothed in water, and (3) draining seedbed. On the last trip over, a spring-tooth har- irrigation water 3 to 5 days after seeding compared row should follow the disk harrow so that prom- with leaving a full flood of water and allowing the inent furrows are left in the seedbed to catch the I'ice to emerge through the water. seed and reduce the drift. He reported slightly higher grain yields when (2) The field should be flooded as rapidly as the seed was soaked to pregerminate it before possible with a minimum of 4 to 6 inches of water seeding in the water. He found that a small per- to control grass. All levees should have a gate or centage of the dry seed floated in the water and spillway so as to make and maintain a constant settled in low spots, whereas the pregerminated level of water and prevent the levees from break- seed remained well in place on both the cloddy and ing because of added pressure from heavy rains. smooth seedbeds. However, he pointed out that (3) The field should be seeded immediately pregerminating seed requires additional labor and RICE IN THE UNITED STATES 109 expense and also requires a short interval between under the conditions of the test, soaking the germination and seeding. seed, muddying up the water, time of removing Faulkner found that the average yield of rice the water after seeding, and depth of the water produced from the smooth or the rough seedbeds up to 8 inches had no effect on the yield of was approximately the same. However, there were water-seeded rice. They felt that it would be a some indications that the working of the soil in good farm practice to leave the surface of the water reduced the germination of grass seed and soil slightly rough in order to lesson the drift thereby reduced the resulting competition. Also he of seed when water seeding rice. found that soils that were worked in the water From observations made on experimental tests dried more quickly when the water was drained and on commercial fields, little if any difference and caused the top one-fourth inch of soil to curl, in suitability to water seeding has been noted thereby pulling the young seedlings from their among varieties. However, in Arkansas where root anchorage. Such conditions would require that the soils are alkaline, or the water contains ex- the area be flushed. cess salts, the medium-grain varieties in general In this series of experiments, Faulkner found have been damaged less from these adverse con- that good stands of rice, along with excellent con- ditions when water seeded than have the long- trol of red rice and grasses, could be obtained by grain varieties, especially the Bluebonnet group. leaving a full flood of Avater on the field until Hall and Thompon {58) pointed out that these the rice had emerged through the water and was problem areas that are colloquially referred to strong enough to stand free from the water. The as alkaline have been brought about, on culti- yield results from 1959 indicated that about 500 vated lands, by the use of well water containing pounds more rice per acre was produced where the excess salts combined with restricted internal water was allowed to remain on the field than drainage of the soil. Some of these troublesome where it was drained a few days after seeding. soils have shown a pH of 7.0 to 7.5, whereas Faulkner thought that on the clay soils it might others have been below 7.0. This indicates that be more difficult to obtain a good stand of rice pH alone is not a completely reliable indicator using the continuous flooding method, but that of such problem soils. additional experiments would be necessary to de- Nelson {97) conducted an investigation to de- termine this. termine the uniformity of distribution of seeds The most common method of land preparation and fertilizers from airplane distributors and for water seeding in Texas in 1954, according to studied the extent to which rate of application, Eeynolds {IIS)^ was plowing and disking to kill materials distributed, and flight altitude affected the vegetation. The land was then left in a uniformity of distribution. He concluded that rough condition until seeding, at which time the a fairly good distribution is possible with well- field was irrigated until the water barely cov- designed equipment if a few precautions are fol- ered the land. It was then harrowed to muddy lowed. Characteristic deposit patterns should the water. Harrowing to muddy the water be- be determined before a new or modified distrib- fore seeding is seldom done early in the season, utor is first used. From the deposit patterns because dry north winds cause the soil surface obtained, optimum flagging intervals can be de- to dry rapidly. This drying curls and cracks termined that will allow for an overlap sufficient the soil surface before the seedlings become to maintain a uniform application or seeding rooted. The seed usually was soaked in bags in rate. For proper overlapping of swaths, a a canal for 24 to 36 hours, removed from the ground crew is necessary. For uniform distribu- water and allowed to drain, and then sown by tion a flagman should be stationed at each end of airplane on the water. Some farmers drained the field, and each flagman should be equipped their fields as soon as possible after sowing, with two long markers connected with a rope or whereas others might delay draining as much as chain length equal to the flagging interval dis- 36 hours. Experiments conducted by Wyche and tance. It is necessary that the flagging interval Cheaney {16It) during 1954 at the Agricultural be measured correctly and that the airplane fly Research and Extension Center showed that directly over the two flags. Recently, many 488-871 0—73- no AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

growers have measured these flagging distances requisite where rice is to be grown. He also em- or intervals and have marked them with stakes phasized the importance of proper use of avail- or small flags on wires so that the flagman will able water. He suggested that under certain con- not need to measure the distance. ditions, it was beneficial to control the flow of fresh water so that it traveled the entire length Transplanting Rice of each cut. This circulation resulted in more Yields of direct-seeded and transplanted rice uniform water temperature and helped to con- were compared in a study conducted by Adair trol algae (scum) and insects. He also stressed and others (7) in the major United States rice the importance of judicious water management areas during a 3-year period starting in 1937. in reducing the effects of certain diseases and other production hazards. Three varieties were included in the experiment at three rice experiment stations in the Southern Amount of Water Required States, and two varieties were included at the The water requirement of the United States rice California Kice Experiment Station. High yields crop is comparatively high because the fields are of transplanted rice often obtained under intense continuously flooded for so much of the growing culture in certain countries had led to a more or season. Rice will not produce a maximum crop on less common belief that yields of transplanted stored soil moisture or infrequent rains, as will rice generally are higher than those from direct other cereals {8). When tried in the United States seeding. Advantages claimed for transplanting or when used elsewhere, the upland system of rice were listed and discussed. It was pointed out culture generally results in yields of only 30 to 70 that high-quality, stiif-strawed varieties already percent of those obtained from flooded rice grown were being grown by machine methods in several under similar soil and climatic conditions. countries. During the 3 years of these tests, the Senewiratne and Mikkelsen {123) suggested that average yields of some of the varieties at some differences in growth responses of flooded and un- locations were significantly higher from direct flooded rice may be due to differences in auxin seeding than from transplanting. None of the metabolism. They found that plants grown under varieties at any of the stations produced signifi- unflooded conditions had a low catalase activity cantly higher average yields when transplanted and a high peroxidase activity, which favored ac- than when directly sown. These workers con- celerated auxin degradation. They suggested that cluded that the principal reasons for transplant- high manganese levels in plants grown under un- ing are not to increase yields but to better use flooded conditions affect the indolacetic oxidase land and labor in densely populated countries in mechanism and result in retarded growth and de- which the tillable land area is limited and the pressed grain yields. Flooded rice grown with labor supply is plentiful. ammonium nitrogen collected small amounts of manganese, whereas plants grown with nitrate ni- Irrigating and Draining trogen (typical of upland rice) contained much Flood irrigation is used for all rice grown in more manganese. Clark, Nearpass, and Specht the United States. The soil is submerged in 2 to {21), however, concluded that "the better growth 8 inches of water most or all of the time from of rice in submerged as compared to upland cul- seeding or shortly after, until the grain is nearly ture in at least some soils is due to greater Mn ripe. This period may extend from 60 to 90 or availability under submerged soil conditions." more days, depending on method of seeding and Jones and others {73) reported that from 2.8 to variety grown (8, 73). Although the water require- 3.8 acre-feet of water were required to produce a ment for rice is high, Jones and others {73) rice crop in the Southern States. About one-third of this is supplied by rainfall during the growing stressed that good surface drainage is as necessary season. Robertson {IH) reported from early ob- for successful rice culture as is a dependable sup- servations that the annual water requirement for ply of good-quality irrigation water. rice in California averaged 8.2 acre-feet but ranged Haskell {5Í) pointed out in 1915 that a plenti- from 4.3 to 14.8 acre-feet. Lourence, Pruitt, and ful and easily accessible water supply is the first Servis {79) reported in 1970 that an average of RICE IN THE UNITED 8TATEÍS 111

2.9 acre-feet of water was required to produce a time. During the next few years, many pumping rice crop at Davis and an average of 3.0 acre-feet plants were installed on the streams in southwest was required at nearby Knights Landing. Louisiana and southeast Texas. By 1901 some On average rice soil, about 1 acre-foot of water pumping plants in operation delivered up to 45,000 is required to prime the soil and flood the paddies gallons of water per minute. to a depth of 6 inches. After a ricefield is flooded, Diesel engines came into common use for pump- a considerable amount of water is required to main- ing water after 1919 (6). Convenience, low labor tain an optimum depth in the field. Water is added requirements, and reasonable initial and operating periodically to compensate for losses due to tran- costs have since caused a shift to electric power and spiration by plants, evaporation from the water natural gas power for irrigating rice. In Arkansas surface, deep percolation, and spillage. These losses in 1955, nearly 50 percent of the 1,800 irrigation will vary, depending on amount of plant growth, installations were powered by electricity. In Lou- solar radiation, temperature, wind, relative humid- isiana, of a total of 1,061 wells, 450 were powered ity, soil type, and rate of inflow of water into the by diesel engines, 212 by natural gas, 105 by elec- field. A rate of flow equal to 1 cubic foot per second tric motors, and the remainder by other units. In (450 gallons per minute) for each 50 acres being some cases water is pumped or relif ted by means of irrigated usually is required to maintain water the power takeoff unit on tractors. levels on ricefields in California {'^8), Historically, most areas in the world where The apparently lower water requirement for water is pumped from wells have experienced de- rice in the Southern States as compared with that clining water tables. This problem, as related to in California is related to naturally occurring cli- Arkansas rice production, was studied by Engler, matic differences between the two areas in the Thompson, and Kazmann (SO). Suggested reme- évapotranspiration losses from ricefields. The rela- dies included the use of deep wells (900 to 1,000 tive humidity during the growing season is quite feet) that normally supply more water than shal- high in the Southern States, but it is comparatively low wells (100 to 150 feet) and the use of small low in California. In California, a vegetation-free reservoirs. Engler (29) reported that during a 30- water surface will lose from 5 to 6 acre-feet in 12 year period, the decline in water level for the months, whereas the corresponding loss in south- Grand Prairie area of Arkansas averaged about ern rice areas may be less than one-half of this three-fourths foot per year, which caused an an- amount. nual decline in capacity of shallow wells of 20 to 25 gallons per minute. Various methods of replen- Source of Water ishing underground water supplies on the farm are discussed by Muckel (92). It is estimated that 30 to 35 percent of the 1962 Gerlow and Mullins (43) pointed out the value rice acreage in the United States was irrigated of small, 20- to 40-acre farm reservoirs in conserv- from wells. Adair and Engler (6) stated that over ing surface runoff water to supplement or replace 40 percent of the 1953 rice acreage was all or partly well water for rice irrigation. Another develop- irrigated from wells. Since that time, however, ment designed to conserve water (whether from numerous additional reservoirs have been con- wells or elsewhere) has been the increasing use of structed—many of them in Arkansas. The reser- underground concrete or Incite irrigation pipelines voirs are filled with runoff water before the (laid about 30 inches beneath the soil surface) to ricegrowing season. Other important sources of transport irrigation water for rice and the accom- irrigation water are rivers, bayous, lakes, and panying rotational crops. Pipeline irrigation drainways. greatly reduces the rodent damage that character- Adair and Engler (6) discussed the early irriga- izes open irrigation systems, and it greatly reduces tion of rice and reported that in 1894 the first large irrigation plant was established near Crowley, La. evaporation and seepage losses. In addition to the After failures of some pumps, a somewhat larger important aspect of water conservation, it has been centrifugal pump was installed in 1896. This pump calculated that, for each 1,000 feet of open canal delivered 5,000 gallons of water per minute, which replaced with a pipeline, 2.8 acres of land are re- was enough for the rice acreage planted at that turned to crop production (S). 112 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

Adair and Engler (6) pointed out that pumping {104^ 107), When soils are strongly saline, having from bayous supplies most of the surface water in an excessively high concentration of sodium, cal- Louisiana and Texas. According to Anderson and cium, or potassium, the concentration of salts in McKie (11)^ sources of rice irrigation water in the soil solution (including the standing water) Mississippi are shallow wells (90 to 100 feet) and may be so great that it will injure or kill the surface water from lakes or streams. Diversion seedling rice (106), Excessive salt concentration from large streams is the main source of water for results in restriction in downward percolation; rice irrigation in California. It is estimated that therefore, the floodwater is subject to a longer less than 5 percent of the California rice acreage period of evaporation with an ensuing increase is irrigated from wells. in salt concentration. Thus, water having a higher salt concentration enters the soil, and the Quality of Water salt concentration of the soil solution is increased. Adair and Engler (6) reported that salt wa- For successful rice production it is very impor- ter put on dry soil damages a ricefield more than tant that the available water be of suitable quality. if salt water is used to replenish the water sup- Rice irrigation water should be relatively free of ply in a field that has been watered with fresh dissolved salts that are toxic to rice plants. The water. The reason given was that the salt was characteristics of irrigation water that determine more concentrated in the dry soil, and more of quality include (1) total concentration of soluble it moved into the root zone, where it was taken salts^ (2) relative proportion of sodium to other up by the plants. They indicated that rice cations, (3) concentration of boron or other toxic grown on clay soils may not be injured as much elements, and (4) under some conditions, the bi- by salt water as rice grown on lighter soils, carbonate concentration as related to the con- because less water is used and less is lost by centration of calcium plus magnesium. Other seepage. important factors to consider are the initial salin- The rice plant can tolerate higher concentra- ity of the soil, the effect of internal drainage on tions of salt in the later stages of growth, al- the flooded soil, and the total salt content of the though very high concentrations may kill the soil. Finfrock and others {S8) described good- plants or make them sterile. Some varieties of quality rice irrigation water thus : rice are more tolerant to salt than others and Specific Electrical Conductivity may make satisfactory yields when the water (K X 10«) lessthan750 contains salt concentrations of 75, 150, 200, Boron, parts per million less than 1 and 250 grains per gallon in the tillering, joint- S.A.R. index (tendency to ing, booting, and heading stage, respectively. It form alkali soil) less than 10.0 is believed that some of the newer varieties would be damaged seriously by such amounts of When high sodium water is regularly used salt (6), each growing season, it may deflocculate the soil, Irrigation water pumped for rice from shal- so that stickiness, compactness, and impermea- low wells in Arkansas and other States fre- bility increase. The deflocculated soil is difficult quently has a comparatively high sodium con- to cultivate and usually produces low yields. tent as compared with water coming from surface Pearson (104,^ IOS) described the point of great- streams. Kapp (76) suggested that the source est importance as the nature of the soil solution and chemical composition of rice irrigation wa- or saturation extract found in the zone of rice ter should be considered from both the imme- roots. If soil saturation extracts have a conduc- diate effect on the current crop and the long- tivity index of 4 to 8 millimhos, the yield of range productivity of the soil. In greenhouse Caloro rice may be reduced 50 percent. and field experiments he found that sodium Bice is very tolerant of salt during germina- chloride added to rice soil injured germination tion, but rice seedlings are very sensitive to salin- and resulted in lowered production of vegetation ity during early development (1 to 2 leaves) and and grain. The addition of 5,700 parts per mil- are progressively less so at 3 to 6 weeks of age lion of sodium chloride to the soil only slightly BICE IN THE' UNITED STATES 113 reduced vegetation, but completely prevented in total salts was used for the remainder of the grain formation. In field trials, 825 pounds of season, and a normal crop of grain was produced. sodium chloride per acre hindered germination, In California, rice culture has proved useful in and 3,300 pounds per acre reduced rice grain reclaiming saline soils, provided the fields to be yield. reclaimed are first engineered to drain well. Water from the shallow wells in Arkansas con- Mackie (81) reported that one rice crop grown taining 75 parts per million calcium and 22 in Imperial clay near Imperial, Calif., reduced parts per million magnesium has caused some the saline content 72 percent to a depth of 6 feet. riceland soils to increase in alkalinity from an In his experiments he found the usual reduction original pH of about 5.0 up to as high as 8.0. in saline content from the first rice crop to be one- The change from a highly acid to a highly al- third to two-thirds. Overstreet and Schulz {103)^ kaline reaction is due to the annual addition of in a series of San Joaquin Valley tests, concluded about 1,500 pounds per acre of limestone equiva- that rice culture serves as an efficient means of lent. The increase in alkalinity has lowered the reclaiming nonsaline soil containing 15 percent or availability of phosphorus in the soil. If a new more of exchangeable sodium. source of water low in dissolved minerals is ob- Water Temperature and Oxygen Content tained, such changes may be reversed (6), Such The temperature of water with which rice is ir- water may be obtained from installing a deep rigated has a profound effect on the plants. Adair well or constructing a reservoir and catching sur- and Engler {6) reported that the temperature of face runoff water. However, as long as the un- rice irrigation water pumped from wells and from favorable soil condition exists, delaying the ini- streams frequently is 65° F. or lower. When such tial flood and following a routine of alternate cool water goes directly into the field, the rice draining and reflooding at 4- or 5-day intervals growing near the w^ater inlet usually is retarded until the plants are about 6 weeks old have given and may ripen as much as 7 to 10 days later than fairly satisfactory relief in Arkansas. the rest of the field. Eaney, Hagan, and Finfrock When the rainfall is below normal in the {111) showed that with the building of high dams Gulf Coast area of Louisiana and Texas, the wa- on the major California streams, the temperature ter level in the streams that supply irrigation of water available for rice irrigation from surface water often is so low that brackish water en- streams had dropped to 51° or less. croaches from the Gulf. The concentration of The temperature of the irrigation water should chloride salts may become so high that the yield be not less than 70° F. nor more than 85° for best and quality of rice is reduced or the crop is results. Raney {110) showed that the critical sea- ruined. Adair and Engler (6) pointed out that sonal threshold of water temperature for normal several workers had shown that water contain- growth of Caloro rice was near 69°. When the ing more than 35 grains of salt per gallon (600 mean temperature was 5° lower, maturity was de- parts per million) should not be used to irrigate layed 30 days beyond the normal 160 days. Rice young rice if the soil is dry and if the water is yield was highest when the mean water tempera- to remain on the field. Eice watered continu- ture was 80°. At water temperatures above 85°, ously with water containing 35 and 75 grains of yield was reduced and root development was poor, salt per gallon w^as reduced in yield about 25 and probably because of low oxygen content of the 70 percent, respectively, and the rice was of lower water. quality than when water containing only 25 grains Chapman and Peterson {20) studied underwater of salt per gallon was used. rice seedling establishment in relation to tempera- Water quality in California ricefields was stud- ture and dissolved oxygen. Under laboratory pot ied by Stromberg and Yamada (147). They culture they found water temperatures in the found that water that contained a high total of range of 77° to 84° F. most favorable for the ef- soluble salts killed rice plants. However, when a fective establishment of rice in static culture. field showing initial symptoms of dying was im- Emergence of the shoot from the water was most mediately flushed out wâth quantities of water low rapid at 84°, but the development of the root sys- in total salts, the plants soon recovered. Water low tem and the penetration of the soil were favored 114 AGRICULTURE HANDBOOK NO. 2 8 9, U.S. DEPT. OF AGRICULTURE

by temperatures as low as 68°. Exposure to a water Water that overñows or is drained from the temperature of 104° for 12 hours or more was le- lowest paddy may be caught in drainage ditches thal to pregerminated Caloro rice seed. They con- or canals and recirculated or moved into other cluded that it seemed unlikely that dissolved oxy- canals or drainage ditches (fig. 34, ^7). Some fields gen deficiency would be a limiting factor in seed- have a canal along one side so that each paddy can ling establishment in the field even at 95°, if cur- be watered separately from this canal. Finfrock rently recommended seeding rates were used and the water had an initial oxygen content of at least 5 to 6 parts per million. Ehrler and Bernstein (ß8) reported that at a constant root temperature of 64.4° F., Caloro shoot gro^vth was twice that at 86° and root growth was one and a half times as great ; however, grain yield was only three-fourths as much at the lower root temperature. No significant interaction was found between root temperature and cationic concentra- tion or cationic rates. Seeds germinate slowly when the temperature is less than 70° F. Water now available for rice irrigation in some areas, including northern Cali- fornia, may be colder than this, ranging down to 51° or less. Such cold water will retard and lower the germination of rice sown in the water and will retard the development of plants so that the stand may be thin and crop maturity will be delayed near the inlet to the field. According to Raney, Hagan, and Finfrock (111), field studies over a 3-year period showed that plant-free water- warming basins, 6 to 12 inches deep, equal in size bo 2 percent of the area to be served, successfully and economically raised the mean water tempera- ture of 60° to 70°.

Water Control Methods

Methods of transporting and controlling irri- gation water for rice production have evolved over the years. They vary in different rice- producing areas but tend to follow a general pattern. In most cases water is conveyed from the pumps, streams, or reservoirs in canals from which it is diverted into laterals, into field c ditches, and finally into the field checks or paddies ife (fig. 34, J. ). Water is usually delivered to the high- PN-2993

est point in the field by canals or pumps. It passes FIGURE 34.—Control of irrigation water: (A), Headgate into successively lower paddies through metal or that controls flow of water from canal into ricefield ; wood levee gates or levee control boxes or openings {B), levee box used in California to regulate flow of in the levees. These gates or boxes provide precise water from one levee to another in ricefleld ; and (C), control of maximum water depth in each paddy pump to reclaim seepage water from ricefield. (All photos courtesy of Agronomy Extension, University of (%.34,5). California, Davis.) BICE IN THE UNITED STATES 115 and others {38) described in detail the rice irri- nia, the levees are usually considerably larger gation control structures used in California. than those in the Southern States. The wood CANAL GATES.—Gates in the irrigation canals or metal boxes are placed along the more acces- of the California ricegrowing areas are simple sible side of the field. Flash boards of various structures of wood or concrete. In the vertical fixed widths are added or removed to regulate the portion of the structure, slots are provided so that water to the desired depth. Boxes must be short planks (flash boards) can be inserted or re- properly constructed and installed so they will moved as desired to raise or lower the water level not be washed out. at the diversion point into the field. Inlet struc- DEPTH STAKES.—To gage the water level in a tures to the field may consist of similar slotted paddy or check, a depth stake is driven into the wooden structures or screwtype metal gates. The soil within a suitable distance of the levee box latter commonly are used in the southern rice or gate. The stake is set so that the bottom of area. an 8-inch red band is at the average elevation LEVEE BOXES OR GATES.—Water is controlled of the ground surface in the paddy. If a 5-inch within the field with boxes or gates set at con- flood is desired, then 3 inches of the red band is venient locations in the levees. In the southern left showing. The stake is painted white at the top, rice area, metal levee gates with adjustable pan- so that if an 8-inch flood is desired, only the white els are in common use in some localities (fig. 35). band is left exposed (fig. 36). Variations that may be used include single planks Construction of access roads around ricefields or boards or metal panels that are forced into helps to (1) cut production costs because equip- the soil across a narrow cut through a levee to ment can be moved more easily into the field, keep the water at the desired level. In Califor- (2) control water better and more economically.

PN-2994 FIGURE 35.—Ricefields in Arkansas being flooded for water-seeding by airplane. Metal gates with adjustable panels are used to regulate water depth in each check or levee area. (Photo courtesy of Arkansas Publicity and Parks Commis- sion, Little, Rock.) 116 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

the chances of breaks. Plastic sheeting or various types of laminated paper also can be used to cover weak areas in levees. Soil is plowed or shoveled onto the edges of the material to prevent its float- ing when the field is flooded. Certain pests, such as muskrats, field mice, Norway rats, crayfish, and large insects, often inhabit canal and drainage ditchbanks, as well as contour levees. Their bur- rowings may result in levee or bank leaks or breaks and heavy water loss. Major structures, such as weirs, can be protected to some extent by trapping or poisoning the pests. In California fields where the large levees sometimes are in- habited by crayfish and muskrats, the problem has been solved by using sheep to graze the vegetation from the levees.

Water Management

The application and management of water is a key operation in rice farming (94). The principal functions of water cover are to control weeds, to condition the soil and provide a favorable environ- ment for rice growth, and, especially in California, to serve as a temperature regulator by minimizing the eñ'ect of the large variations between daytime and nighttime air temperatures. PX-2995 FIGURE 36.—Depth stake used to determine depth of irri- Systems of water management for rice produc- gation water in ricefield. Nails are spaced 1 inch apart. tion vary widely, depending on method of seeding, (Photo courtesy of Agronomy Extension, university of soil type, climate, crop rotation, diseases, and in- California, Davis.) sects. A very important aspect of water manage- ment is good drainage. Tliis includes drainage of and (3) control mosquitoes by reducing seepage winter rainwater as well as periodic drainage dur- problems, by making it easier to control weeds ing the growing season. In years when riceland where mosquitoes breed, and by enabling mos- is used for rotational crops, the crops may be quito-abatement employees to reach troublesome drowned if drainage is poor. Seedbed preparation spots. and fall harvest can be expedited by a good drain- PROTECTION or IRRIGATION SYSTEM.—Once they age system. Drainways around and through the are engineered and installed, riceland irrigation field that connect with main drainage ditches lead- and drainage structures are subject to damage ing to natural drainage channels are essential. from use, the elements, and insect or animal pests; These ditches should be large enough and deep but countermeasures can be employed. In large, enough to allow the removal of large quantities flat paddies or checks, ^\ind levees often are con- of water quickly. When it is necessary to preirri- structed between normal-interval levees to decrease gate or to flush irrigate to bring about emergence wave action. Their judicious placement in areas of rice, good drainage plays a key role in getting subject to strong winds may be very beneficial. So- a good stand quickly by expediting fieldwork and called alkali levees built from soil high in sodium reducing seed rot losses (8, 73). Good drainage of and other salts fi-equently are subject to washouts ricefields and adjacent areas and ditches also is of or breaks wlien fields are flooded. Heavy applica- prime importance in eliminating mosquito-breed- tions of gj'psum on such areas before pulling the ing areas. levees help to stabilize the soil and greatly reduce Essentially, there are two broad systems of RICE IN THE UNITED STATES 117 water management, depending on the method of or it may be delayed for 2 or 3 weeks. The major- seeding. The first system revolves around drilling ity of ricegrowers in Arkansas formerly drained or broadcast seeding on "dry" ground. Where nec- their fields about midway in the growth of the rice essary, seeding is followed by ñushing to bring and allowed the soil to dry before applying nitro- about uniform emergence, and the first flood is gen fertilizer ; then they reflooded the fields. applied later. This system is often used in the Eesults from research conducted at Stuttgart, southern rice area. With the second broad system Ark., have brought about major changes in water of water management, fields are flooded just be- management. Many ricegrowers no longer drain fore aerial seeding and usually remain flooded their fields at midseason unless they have some spe- until they are drained for harvest. This system is cial reason for doing so. Currently, the common used in California, and with modification also is practice on drill-seeded rice is to apply herbicide used in the southern rice area. In the southern rice for grass control when barnyardgress is in the 1- to area (especially in Texas), where rice is water 3-leaf stage of growth. This usually is about 8 to 12 seeded on heavy soils, some growers drain as soon days after the rice seedlings emerge. About 2 days as possible after seeding; others may delay drain- later nitrogen fertilizer (about 40 to 50 percent ing for 36 hours. After stand establishment, irri- of the total amount to be used) is applied by air- gation practices are essentially the same as for plane, and the field is flooded immediately. Unless the field must be drained for midseason disease, drilled rice {8), insect, or weed control, a flood often is maintained Unless rice is water seeded and the flood is maintained, a flood is applied about as soon as the continuously until the field is drained about 10 to crop is old enough to withstand submergence. Be- 15 days before harvest. An additional amount of nitrogen fertilizer usu- fore the use of chemicals for weedy grass control, ally is applied near midseason, using culm elonga- ricefields on the lighter soils in Arkansas often tion as a guide to optimum timing (5^, 129), This were flooded as soon as the rice had emerged. If method of timing the application of nitrogen re- the fields were free of weedy grasses or if these grasses had been chemically controlled, the land sults in shorter plant height, less lodging, and in- might not be submerged until the rice seedlings creased grain yield. As an example of the impor- tance of proper timing, Johnston and others (70) were 4 to 6 inches tall. Jones and others {7S) indi- reported that after applying 40 pounds of nitro- cated that at this stage the land is submerged to a gen per acre to the Nova 66 variety just before the depth of 2 to 4 inches. During the rest of the grow- first flood, delaying application of the other 40 ing season or until the land is drained before har- pounds of nitrogen from 43 until 67 days after vesting (except for special reasons), the water can seedling emergence resulted in shorter plant height be held on the land at this depth. To maintain this constant depth, additional water should be applied (from 54 to only 47 inches), less lodging (from 69 to replace that lost by evaporation, transpiration, to only 2 percent), and increased grain yield (from 5,063 to 7,059 pounds per acre). Little, if any, in- and seepage. ternode elongation had occurred at 43 days; at 67 Special reasons for draining ricefields during the longest internodes averaged about 38 milli- the growing season and allowing the soil to dry meters. include (1) control of algae (scum), (2) control The midseason or "internode" application of of rice water weevil, (3) prevention of straight- nitrogen fertilizer usually is made when a shallow head (blight), and (4) control of aquatic weeds flood covers the field. No additional water should such as ducksalad. On saline or alkaline soils, fields be added to the field until at least 1 day after the may need to be flushed or drained several times early in the growth of the rice to allow the plants nitrogen is applied. When relatively high rates to become well established. of nitrogen are used on stiff-straw varieties, a sec- ARKANSAS.—Adair, Miller, and Beachell (8) re- ond midseason application can be made 10 to 14 ported that water management is similar in Ar- days after the first midseason application. In such kansas and Louisiana. Depending on growing con- cases, 25 to 30 percent of the total amount of ni- ditions and grass control methods being used, the trogen is applied in each of the midseason applica- first flood may be applied as the rice is emerging. tion. Detailed recommendations are presented by 118 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE.

Huey and Chapman {62). They specify the criti- it is 6 to 8 inches tall, and then only to a depth cal average length of the first elongated intemode of 41/è inches. Fields are drained as necessary at which the first midseason application of nitro- for topdressing with fertilizer and for pest gen should be made to each variety. control. A water-seeding method formerly used by some MISSISSIPPI.—Mullins {91^) stated that rice in Arkansas growers {51) has been modified by re- the delta area of Mississippi is kept flooded for cent results. For the new method, a suitable seed- 90 to 120 days during the growing season, depend- bed is prepared and immediately after flooding, ing on variety and seeding date. The first flood seed is broadcast by plane into fields. On silt loam is applied as early as possible after the rice seed or other relatively light soils, the flood can be held germinates and a stand is established. This may for about 10 days and then drained for application be about 2 weeks after seeding or, if germination of early-season herbicide. About 2 days after herbi- is slow, possibly 3 weeks after seeding. Applica- cide treatment, nitrogen fertilizer is applied and tion of the first flood requires close attention to the field is reflooded immediately. Unless some spe- avoid breakage of levees and to stabilize the water cial need for draining arises, the field is kept at the desired level. If the levees are accurately flooded until about 10 days before harvest. When located and well constructed, relatively little time rice is water seeded on clay soils, the fields are is required to apply the first water; otherwise, drained soon after the seed sprouts and are not re- labor requirements may be much higher. Mullins flooded until after herbicide and nitrogen have reported that most fields were drained after 3 to been applied. If water-seeded fields are left drained 4 weeks under flood and the soil was allowed to for several days, heavy stands of grassy weeds may dry for several days. This permitted the roots become established and additional early-season of the rice plant to become more firmly estab- chemical control will be required to obtain and lished and also stimulated additional tillering. maintain satisfactory stands of rice. He suggested that if grass was well under control Some ricegrowers in Louisiana, Mississippi, and at this time, fertilizer could then be applied. Texas report using many of the same water and TEXAS.—Reynolds {113) stated that where rice fertilizer management practices as those described is seeded with a grain drill on heavy soils, the for ricegrowers in Arkansas. fields usually are flushed (irrigated) for germina- In a series of greenhouse experiments, Hall tion if water is available and if the soil has not {50) found that unless rice plants were fertilized, been saturated by rain. Flushing usually is the practice of draining and drying the soil and practiced in areas irrigated from canals, since the then reflooding did not increase grain produc- fields can be covered rapidly. Sandy soils and tion on soil low in organic matter. soils irrigated from wells usually are not flushed. LOUISIANA.—Jenkins and Jones {65) found After fields are flooded, irrigation water may be that in date-of-submergence experiments in Loui- drained off once or twice during the growing siana, the highest average yields were obtained on season to permit fertilization and to control water land submerged 20 days after the seedlings had weeds and insects. The time and number of emerged. In a discontinuous and continuous sub- drainings may vary, depending on the length of mergence experiment, early continuous submerg- maturity of the variety, the presence of weeds ence (10 days after seedling emergence) of the and insects, and the supply of irrigation water. land gave higher average yields than did inter- Where there is a shortage of water for reflooding, mittent drying of the land followed by continuous tlie fields usually are not drained. If irrigation submergence. is required to germinate the seed, the field is In Louisiana, water seeding is not common ; but promptly drained after this flushing. Where rice where it is practiced, the water is drained when is seeded with an endgate seeder, the land is then the rice seedlings are one-half inch long, accord- liarrowed and irrigated and the irrigation water ing to Wasson and Walker {157). It then is is drained off shortly thereafter. It has been allowed to grow until flooding is needed. Drilled found from tests and general observations that rice may be flushed if necessary for uniform ger- satisfactory rice stands are not obtained if seeds mination. Normally the rice is not flooded until are covered by both soil and water, so that drain- RICE IN THE' UNITED STATES 119 ing water from fields where the seed is covered Relative fertility and varietal responses were simi- with soil is extremely important. lar for all water management systems. Daytime Morrison {89) found little difference in the rice water temperature, tillers per plant, shoot dry yields at Beaumont when the total amount of weight, plant population, active leaf area per water used ranged from 46 to 73 inches. He con- plant, panicles per plant, panicles per square cluded that the use of 45 to 50 inches of water meter, total nitrogen in the shoots at 30 days, and would be just as satisfactory as the use of larger total nitrogen in the grain all were greater in shal- amounts. These results generally agree with those low water than in the other water management reported by Jones and others (7^). It was further systems. Seedling emergence and flowering were concluded {89) that the depth of water does not earlier in shallow water, but lodging was greater seem to be of much importance where weeds are and total nitrogen in the straw was lower than not a problem. Draining ricefields once during the in the other systems. The greater number of pani- growing season increased the yield of rice con- cles per square meter was the yield component that contributed most significantly to the higher yields siderably. Evatt {31) reported that preliminary tests con- obtained with shallow water. Water can be low- ducted at the Eice-Pasture Research and Exten- ered to about a 1- to 2-inch depth for weed control and at stand-establishment time during cool sion Center in 1956 and 1957 showed significantly weather. If the soil in the field contains excessive reduced rough rice yields of the Century Patna 231 amounts of salts, the water may become sufficiently variety with use of deep water (10 to 12 inches), compared with use of average depths of water of saline as to be toxic to the rice seedlings. In such from 4 to 6 inches from late tillering to maturity. cases the water should be drained from the field He found that the temperature variation was 2 to and fresh water applied within a few days. 4° F. greater under the shallow water than under Draining for Harvest the deep water. Yields from superimposed fer- tilizer factorials showed no significant fertilizer- Draining at the proper time before harvest is water depth interactions in 1956. However, in necessary to dry the soil enough to support har- 1957 yields were less from both nitrogen and vesting equipment. It is equally important to hold phosphorus treatments under the deeper water. the water on the land long enough to permit the CALIFORNIA.—Before selective herbicides were rice to reach proper maturity. available for the control of grassy weeds, the con- The time to drain depends on the type of soil, tinuous flood system commonly used in California drainage facilities, and seasonal weather condi- was developed to control weeds, principally water- tions. Some soils dry and crust quickly after grass, and to fit the fertilizer requirements of rice drainage; others dry slowly. Less time is required {39, 87). Ricefields normally were rapidly flooded to dry the soil early in the season when tempera- to a depth of 6 to 8 inches before seeding and tures are higher and the days are longer than is re- then seeded immediately with presoaked seed. The quired later in the fall. Growers soon familiarize water was maintained at that depth until it was themselves with the drying time of their soil, so drained about 30 days before harvest. With the that they can judge when to drain the fields to per- advent of a series of new chemicals for barnyard- mit harvesting and yet not let the crop suffer for grass control, many growers have turned to the lack of soil moisture. shallow (2 to 4 inches) water system. Oelke and Usually the land may be drained when the rice Mueller {102) provided the research findings that is fully headed and the heads are turned down triggered adoption of the shallow water system. and are ripening in the upper parts. This stage Several water-management systems for water- ordinarily will be about 2 to 3 weeks before the seeded rice, each with several nitrogen levels, were crop is ready to cut. The date when the rice will compared for 3 years to determine their influence be ready for harvest can be estimated by observ- on growth and yield. A shallow (1.6 inches) water ing the date of first heading (when approxi- depth all season consistently gave higher yields mately one-tenth of the rice heads have emerged). than intermediate (3.2 inches), deep (7 inches), or Rice in the South normally requires 35 to 45 days fluctuating (1.6 and then 7 inches) water depths. from first heading to maturity. In California, 120 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

crops of average yield will be ready for harvest to ricefield conditions. Low wet spots, weed about 45 days from the first heading. Heavy crops patches, and high, narrow levees made combine with yields of 5,000 pounds or more per acre may harvesting difficult. require as many as 55 days from first heading to At the same time that combine harvesting was maturity. first being tried, tests were being conducted on Water intake should be discontinued a few days artificial drying of rice {137). Even with the prob- before final drainage. The water already on the lems encountered in combining and artificial dry- field will then recede slowly. This lessens lodging ing of rice, numerous advantages were cited. These and does not overtax the drainage system with included: (1) In one operation rice is removed excessive water. In the South the field may be from fields with no danger of weather damage; (2) drained by removing the levee gate panels or by very little loss of rice is sustained through shatter- "cutting'' the comparatively small levees with ing; (3) rice of high milling quality can be obtain- hand shovels. Levee gates sometimes are upended ed regardless of weather conditions at harvest; and or may be removed from the fields before harvest. (4) all the rice can be thoroughly and uniformly Dynamite is used quite extensively in California dried, making it safe for storage either in bulk or for opening levees to permit draining of ricefields in sacks. Also, it was pointed out that artificially before harvest. The levees are blown where they dried rice usually is more uniform in quality be- cross the drains that are installed within the fields. cause of the thorough mixing during the drying The dynamiting, which requires experienced pow- operations. dermen, is started at the lower end of the field. Additional advantages of the combine method over the binder, as pointed out by Slusher and Harvesting, Drying, and Storing Mullins {132) were (1) the elimination, to a large extent, of the need for outside labor for harvest; Harvesting (2) less chance for loss from destructive wildlife, Most rice grown in the United States now is di- principally blackbirds and ducks; (3) possibility rectly harvested with self-propelled combines be- of salvaging a much higher percentage of crop in cause this is the most efficient and economic method. case of storm or %vind damage ; and (4) reportedly The rice is then dried artificially before it is stored less field loss. These authors found that hand labor or milled. Careful adjustment of the combine and used in harvesting was reduced from slightly more proper drying methods result in grain having high- than 11 hours per acre with the binder to from 2.4 est milling quality and commercial value. to 3.6 hours with the combine method, depending Before the adoption of direct combining, the on the size of the machine used. They reported that United States rice industry used three other har- 400 acres of rice approaches the maximum sea- vest methods, according to Smith {185), These, sonal acreage that a grower can safely plan on in order, were: (1) Harvesting by hand with a harvesting with one 12-foot self-propelled com- sickle or cradle, and then stationary threshing; bine. Slusher {131) indicated that the usual crew (2) cutting with team-drawn or ultimately with for the self-propelled combine method consisted of tractor-drawn grain binders, handshocking the one man on the combine, one man with a tractor bundles, and then engine-powered stationary and grain cart to haul grain from the combine to threshing; and (3) cutting with a tractor-drawn the truck, and two men and trucks to haul grain to header or swather, followed by threshing from tlie drier or elevator. One such crew usually could the windrow (when the grain was dry) with a harvest about 16 acres of rice per 10-hour day. pickup combine. Mullins {94) emphasized that the time required Direct combining was first tested in Arkansas and the cost of harvesting depend largely on the and Texas in 1929 {137), According to Bainer weather at harvestime. Heavy rains and wind {IS), approximately 3,000 acres of rice were har- often cause excessive lodging and create unfavor- vested by combines in 1929 in California, and able surface conditions for operating combines and nearly 35,000 acres in 1931. The combined har- other equipment. Severely lodged rice may be vester-thresher had been developed primarily for harvested with combines that have adjustable wheat, and the first models were not well suited pickup reels. When this is necessary, the combine RICE IN THE' UNITED STATES 121 must be operated at a much lower speed across a Hurst and Humphries (63) discussed harvest- field, which greatly increases the time required for ing of rice with combines and pointed out that harvesting. care should be taken not to crowd the feed. Because The variety of rice being grown and proper rice straw is heavy and green at harvest, the fertilizer management are important factors in the ground speed of the machine should be slow amount and type of lodging that may occur. Field enough to avoid clogging. Since the rice kernel is observations confirmed by grower reports indicate very susceptible to cracking, the cylinders should that varieties may differ noticeably in the type of be run at a slower speed than for other cereal lodging. This is particularly true in the case of the crops. Allowing a small percentage of kernels to medium-grain varieties Nato and Nova 66. When crack may be necessary to thresh out a maximum lodging is rather severe, Nato plants characteris- yield. tically fall over at the ground line so that the stems Special self-propelled rice combines used in are nearly flat against the ground and combine California are equipped with crawler tracks that harvesting is difficult and slow. In contrast, the enable the machine to cross wet spots, small stems of Nova 66 usually bend over about 8 to 10 ditches, and low levees. In rainy seasons, wide inches above the ground and such fields can be wooden mud cleats are bolted on to the tracks to harvested much more rapidly. Growing new, increase the support for the harvester. Self-pro- shorter and stiffer straw varieties such as Star- pelled combines used in the ricefields in the South bonnet under proper fertilizer management can usually are equipped with large tires with mud be an important factor in reducing harvesting lugs so that they can be operated over the sloping costs. Because Starbonnet produces less vegeta- levees. tive growth than Bluebonnet 50, less straw is In recent years, specially designed rice combines taken into the combine, and fields can be har- have been built by the major farm machinery man- vested in a shorter period of time. However, this ufacturers in the United States and are being can create a problem such as that experienced used in the ricefields of the South and the West. These high-powered machines commonly have 12- in Arkansas in 1969. In central Arkansas, a to 16-foot headers and can be operated under ex- large percentage of the rice acreage was seeded to tremely muddy conditions. They can be adjusted Starbonnet and most of this acreage was seeded in to do a thorough job of threshing with a minimum a period of about 7 to 10 days. In the fall most of of shelling and cracking of the grain. Most com- this large acreage also matured in a very short pe- bines now are equipped with straw spreaders or riod of time and extremely large numbers of trucks choppers. The latter cut up the rice straw as it had to wait in line before they could unload at the leaves the combine, and the straw particles are driers. Because of the congestion, some driers were spread uniformly over the stubble to facilitate unable to accept deliveries for 2- to 3-day periods plowing under. of time. Where large acreages are seeded to a given The effect of combine adjustment on harvest variety, seeding dates should be spread out in or- losses of rice was studied by McNeal (83). He der to prevent such congestion at driers. found that there were four types of combine Most rice now is handled in bulk (133), al- losses—by cutter bar, cylinder, rack, and shoe. To though some seed rice is sack dried. Self-propelled obtain maximum grain yields, it was necessary for rice combine harvesters are equipped with rela- the cylinder bars and the concave to be in good tively large bins or hoppers for collecting the condition and for the concaves and other parts threshed grain (fig. 37, J.). The hoppers are emp- of the combine to be properly adjusted. He con- tied by mechanically augering the rice into self- cluded that combine ground speed should be re- propelled "bankouts" or tractor-drawn carts that duced to one-half mile per hour when the rice take the rice to waiting field-side trucks (fig. 37, is badly lodged. He indicated that the operator B), Kice then is hauled to driers (figs. 37, O and is the most important factor in preventing high 37, D) or to aeration bins (fig. 38) where it is un- combine losses. The height at which the rice was loaded by use of grain augers or other bulk-han- cut was very important, since the proper amount dling methods. of straw served as a cushion to the grain in the 122 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE.

threshing process, and resulted in a lower cylin- Their tests in rice indicated that total combine der loss and less hauling and breakage. loss including reel shattering should be less than Curley and Goss {22) found that combine 5 percent of the gross yield if the machine is losses also may be due to overloading or improper properly adjusted and operated. Thus, the loss machine adjustment or a combination of the two. from a 6,000 pound per acre yield should not Overloading as a result of excessive ground speed exceed 300 pounds. A high loss of unthreshed usually is the major cause for excessive loss in all grain usually resulted from improper cylinder or sizes of combines. concave adjustments, or from both. High loss of

^"-^"55 BN-22016 FIGURE 37.—(A), Cutting and threshing rice with a combine harvester. One combine is emptying its hopper bin into "bank-out," or grain cart ; (ß) transferring rice from the "bank-out" into truck in road at fleldside ; (C) trucks loaded with rice waiting to unload at farm drier ; (D) unloading rice at drier RICE IN THE UNITED STATES 123

BN-22015

FIGURE 38.—(A), Bin drying combined rice on farm; (B), farmer cooperative drying and storage facility. Fans aerate rice in flat bins. Silo storage in center and column drier in rear. (Photo for B courtesy of Agronomy Extension, University of California, Davis.)

threshed seed came from poor separation in the portions of the panicle and in the hard-dough straw walkers and cleaning shoe. They found stage at the base. Few, if any, chalky kernels were that under California conditions reel shattering found in rice harvested at this stage of maturity. losses normally were insignificant in rice as com- When such rice was properly dried, germination pared with other small grains. was satisfactory for seed purposes. Curley and Goss suggest that several loss checks Bainer (13), Davis (£5), Kester (76), McNeal should be made on a machine in a given area to (84.), Morse and others (91), and Smith and others determine the effect of adjustment or change in {136, 137, 138) have reported results of research ground speed. Detailed recommendations for ma- relating stage of maturity (moisture content) of chine and procedural adjustments for harvesting the grain to proper harvesttime. Although their rice under California conditions also are listed by findings varied, they were in general agreement these workers. Electronic harvester attachments that maximum yield of head rice was obtained now are available to monitor grain losses going when rice was harvested at a moisture content of into the threshed straw. about 18 to 24 percent and then immediately dried MOISTURE CONTEXT OF GRAIN AT HARVEST.— to between 13 and 14 percent. Smith and others (138) pointed out that rice must Varieties differ in the range of harvesttime be of high milling quality to command a premium moisture content at which they yield the best price, and that for this high quality and for quality milled rice. McNeal (8^) reported that maximum grain yields, rice must be cut at the a range in moisture content of 16 to 22 percent proper stage of maturity. If the crop is harvested provided the highest yield of head rice for Rex- when immature, field yields usually are redu>_ed ark and that a range of 17 to 23 percent provided and the breakage in combining and milling is the highest yield for Zenith. Davis {'25) reported excessive because of the light, chalky kernels. If that the best range for Caloro was 20 to 25 per- the crop is left in the field until overripe, the cent, and Kester {76) and Kester and Pence {77) kernels may check. This causes severe breakage reported that the best range for Calrose was 22 to during combining and milling and a reduc- 27 percent. Their work showed that for every tion in the yield of head rice (whole kernels). 1-percent reduction in kernel moisture while the Smith and others emphasized that when rice has unharvested grain was left in the field, the decline reached the proper stage, harvesting should pro- i; head rice yield from the maximum was 1.4 per- ceed rapidly, since loss of moisture in standing rice cent for Caloro, 0.9 percent for Calrose, and 1.0 may be very rapid. In rice harvested at the proper percent for Colusa. stage, the grains were fully mature in the upper Most ricegrowers determine the moisture con- 124 AGRICULTURE HANDBOOK NO. 2 8 9, U.S. DEPT. OF AGRICULTURE tent of hand-harvested samples of their rice be- drying of seed rice. However, no grower should fore beginning harvest. Determinations usually use a desiccating chemical on his maturing rice are made with electrically operated moisture crop until he has checked with his local agricul- meters, which they own or which are available at tural authorities to determine its legal status with the commercial driers. reference to chemical residue tolerances. Laws Although kernel moisture content at harvest- are strictly enforced in this regard, and an entire time strongly influences rice quality and milling crop could be impounded if it exceeds the legally characters, environmental factors that affect the established chemical residue tolerances. plant physiologically during the growing season also influence quality. Halick {Jf9) has shown that Drying and Storing a variety that ripened under lower temperatures For the proper drying of rice, moisture must be (81° to 84° F.) produced higher head rice milling removed from inside the kernel (^^). If rice is yields than did rice that ripened earlier in the dried too rapidly or if the temperature of the season under higher temperatures (90° to 94°). drying air is too high, quality is seriously im- Stansel, Halick, and Kramer {lJf5) found that paired. To prevent internal checking or breaking high temperatures (90° day and 80° night) in- of the kernels from drying too rapidly, drying creased chalkiness in all four varieties studied usually is done in three to five stages. In each (Century Patna 231, Bluebonnet 50, Toro, and stage the rice passes through the drier and then a glutinous variety). California rice quality is tempered in a bin, so that the kernel moisture studies {77) showed that certain crop fertilization will equilibrate. practices could result in declines in kernel char- Bainer {IS) published one of the earliest ac- acters such as w^eight, water uptake, hot-paste counts of artificial drying of combine-harvested viscosity, and whiteness. Thus, when harvesting rice. He concluded that rice could be dried suc- for maximum quality, many factors must be con- cessfully by artificial means but that the tem- sidered; but certainly moisture content of grain peratures in the drier should not exceed 100° F. at harvesttime is among the most important. Smith and others {137) reported on early re- PREHARVEST CHEMICAL DRYING.—Eesults from search on artificial drying of rice in Arkansas and experiments with the preharvest application of Texas. They concluded that a drying-air tem- chemical desiccants to speed the drying of rice in perature of 120° F. could be used without injur- the field have been reported by Addicott and Lynch (P), Hinkle {59, 60, 61), Smith, Hinkle, ing the rice if the moisture content was reduced only about 2 percent at each drying operation and and Williams {ISJf), Tullis {151), and Williams {160). All materials were applied as sprays when the rice was allowed to remain in storage 12 to 24 hours between drying periods. However, when the rice contained from 20 to 27 percent mois- necessary to dry a given lot of rice in one opera- ture. Materials tested at varying concentrations included (1) sodium chlorate-borate mixtures; tion, the drying-air temperature should not exceed (2) magnesium chlorate; (3) disodium 3,6- 110°. endoxohexahydrophthalate (alone and with am- From experiments conducted with combined monium sulfate) ; (4) sodium monochloroacetate ; rice in 1944, 1945, and 1946, McNeal {82) con- (5) S, S, S,-tributyl phosphorotrithioate; (6) cluded that in most cases the head rice yield was sodium pentachlorophenate) ; (7) di-nitro-0-sec- increased and the total drying time was decreased butylphenal compounds (alone and in combina- as the number of dryings was increased from one tion with aromatic oils) ; and (8) aromatic oils to four at temperatures ranging from 100° to alone. Some of the chemicals studied by these 150° F.; and the tempering period between dry- investigators hastened the drying of rice in the ings was important, since it gave the moisture in field. However, none was entirely satisfactory the grain an opportunity to equalize and thereby because the milling quality was reduced, the ker- reduce drying time. On the basis of more recent nels were discolored, or the chemical imparted an experiments, McNeal {85) concluded that the off-flavor to the rice. most desirable drying combination appeared to be Desiccants are used occasionally to hasten the four dryings at 120°. RICE IN THE UNITED iSTATES 125

In 1953, Aldred {10) summarized research on bins usually at depths of 6 to 8 feet. Normally drying and storing rough rice in the Southern the grain is dried in the same bin in which it is States. This summary covered work conducted stored, which makes the method particularly by the Agricultural Experiment Stations in suited to on-farm installations. In contrast, aera- Arkansas, Louisiana, and Texas, and by the U.S. tion is defined as the procedure used to cool and Department of Agriculture. This author gives ventilate grain during storage to maintain qual- information on volume of air used in column- ity. This is accomplished by turning the grain type and bin-type driers; number of stages to at frequent intervals, by transferring the grain use in drying rice; temperature of the air for from one bin to another, or by circulating air drying ; depth of rice when dried in bins ; length through the stored grain. Although drying may of time rice can be held before drying; and the be accomplished with unheated (normal atmos- use of fumigants to protect rice from storage pheric) air under favorable conditions, a source insects. of supplemental heat should be available during A comprehensive review of research in the periods of high humidity. United States on conditioning and storing rough Bin drying, sometimes consisting of drying with and milled rice through 1958 {23) was published unheated air but often supplemented with artifi- in 1959. This review cited 74 specific references cial heat, is used in on-farm installations, espe- and included participation by 36 contributors. cially for seed rice. Supplemental heat refers to TYPES OF DRIERS.—Driers can be classified by heat added to atmospheric air for limited tempera- types as (1) continuous fiow, including mixing and ture rise, usually less than 20° F., to accomplish nonmixing; and (2) batch, including bins and drying within a maximum permissible time to pre- potholes. Driers can also be classified as multi- vent spoilage. Barr and others {15)^ Henderson pass and uni-pass. Multi-pass is a continuous- (•55, 56^ 57)^ Hildreth and Sorenson (5<9), Soren- flow system; uni-pass is a batch system. son {IJfO^ HI)') and Sorenson and Davis {IJß)^ all Wasserman and others {loi) reported that discuss the requirements for successful bin drying mixing-type driers are of many designs. Two of and its limitations. the most popular mixing-type driers are the baffle Sorenson, Davis, and Hollingsworth {IJ^) design and the Louisiana State University (LSU) pointed out that some rice for seed is dried in pot- design (i). In the baffle design, rice is conducted hole-type sack driers. downward in a zigzag path by means of baffles, McNeal {85) describes two distinct procedures while heated air is forced through the grain. In in use for commercial rice drying. These are the the LSU design of the baffle drier, layers of multi-pass system and the uni-pass system. In the inverted-V-shaped air channels are installed in multi-pass driers, air heated to approximately 120° a bin with alternating air-inlet and air-exhaust F. is passed through a continuously moving col- channels. Each layer is oiïset, so that the tops of umn of rice in three or more passes. During the the inverted V's split the streams of grain as it first pass a predetermined amount of moisture is flows down between the channels. Drying air removed ; then the rice is stored for about 24 hours passes from the air-inlet channels through the rice to allow the moisture to equilibrate throughout the and out the air-exhaust channels {15Jf). individual kernels. As many additional passes are Wasserman and others {ISJt) stated that the non- used as are necessary to reduce the moisture con- mixing columnar type drier is the simplest and tent of the rice to a safe storage level. most commonly used. In this drier, rice descends In uni-pass driers, air heated to 110° F. or between two parallel screens set 4 to 6 inches apart higher is passed continuously through the column while heated air is blown through the screens and of rice in one operation until the moisture content intervening rice. No appreciable mixing occurs, of the grain is reduced to safe storage level. This and the effect is similar to drying in a static bed system employs a pack 10 inches thick, and the with a depth equal to the distance between the direction of the heated drying air is reversed at screens. regular intervals and the drying is accomplished According to Morrison, Davis, and Sorenson in one pass. {90)^ bin drying refers to drying grain in storage SOURCES OF HEAT FOR DRIERS.—^The amount of 126 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

heat required for warming the air used by a rice the viability of naturally moist rough rice were drier depends on the initial temperature and mois- much the same as for other kinds of seed similarly ture content of the air. The approximate quantity treated. General observations included the follow- of heat required to raise the temperature of each ing: (1) For a given sample of rice, the zone of 1,000 cubic feet of air and its moisture 1° F. is heat damage (beginning of heat damage to loss of about 18.1 B.t.u.'s. To take care of initial air viability) is narrow; (2) the resistance of rough temperature, heat losses, and periods of above rice to impairment of its viability by heat varies average humidity, results of experiments at inversely as its moisture content; and (3) when Crowley, La., indicated that the heat source viability of rice is damaged by heat, loss in germi- should provide 2 B.t.u.'s per minute for each cubic nation vigor is apparent before loss in germination foot of air used. The most satisfactory source of capacity. heat was found to be natural or butane gas. It was In artificial drying experiments with seed rice, pointed out that gas burners are relatively inex- good results were obtained by continually rotat- pensive and are easily regulated to maintain a ing the seed from bin to drier to another bin until desired air temperature. Also, since a clean flame dry. An inlet air temperature of 120° F. and a is produced, the burner can be directed toward the 20-minute drying period were used. This method inlet to the blower and the products of combustion reduced drier capacity somewhat but eliminated allowed to pass through the drier (1). the possibility of mold or fungus attacking the DRYING EATE AND TEMPERATURE.—^Unlike most germ of warm, damp rice between drying periods cereals, rice is consumed primarily as unbroken (1). McNeal (82) showed that germination per- kernels, so that the market value for whole ker- centages were reduced when rice was dried at air nels is much greater than that for broken kernels. temperatures above 130°. Therefore, to avoid breaking the rice kernels, IMPROVING EFFICIENCY OF COMMERCIAL much more care is required in drying rice than in DRIERS.—Recent studies have shown how commer- drying other cereals. The drying rate of rice de- cial drier capacity could be increased by about 50 pends on the rate of migration of moisture from percent and head rice yield by 2.5 percent without the inside to the outside of the kernel. As the rice significant additional capital outlay or increased kernel is dried, the outer portion shrinks, setting operating costs {109^ 144,155,156), Depending on up stresses and strains. When too much moisture the physical characteristics of the drying equip- is removed too rapidly, checking or shattering of ment, a combination of increased rate of flow the kernel results. To reduce this elffect, rice is through the driers (up to 50 percent) and higher dried with air at 100° to 130° F. in two or more drying temperatures (up to 150° F. for Caloro stages or passes through the drier. Between passes rice) is used to achieve increased drier efficiency. through the drier, the grain is held in a bin to allow The findings are based on tests with rice varie- the moisture to equilibrate throughout the in- ties in southern areas as well as in California. dividual kernels. This tempering relieves stresses Calderwood and Hutchinson {17) conducted a and strains and facilitates drying in the next pass. series of experiments at Beaumont, Tex., to de- Laboratory-scale studies with California-grown termine how best to hold freshly harvested rice rice showed how drying-air temperature and num- until it could be dried. These studies, as well as ber of passes through the drier affected head rice those of Hutchinson and Willms {64)-, also in- yield and total drying time. The relation of these cluded the use of unheated air to effect economies factors has been represented on a single diagram in rapidly cooling rice during the tempering to serve as an operating guide in the drying of periods between passes in the heated air column rice (154^). driers. Long-grain rice, with an initial moisture EFFECTS OF HIGH TEMPERATURES ON SEED VIA- content below 20 percent, aerated with 0.43 cubic BILITY.—McFarlane, Hogan, and McLemore (80) feet per minute per 100 pounds, showed no change studied the effects of heat treatment on the via- in grade when held in temporary aerated storage bility of rice. They also surveyed the literature for 10 days before drying. Medium-grain Gulf- involving rice and other grains. Germination tests rose, with an initial moisture content of 23 per- confirmed that the effects of high temperatures on cent, aerated with 0.43 cubic feet per minute per RICE IN THE UNITED STATEiS 127

100 pounds, was not damaged when stored for 3 tion against rain or flood damage. Offsetting these days. Another lot, with an initial moisture con- are usually lower cash outlays per unit volume of tent of 22 percent, aerated with 1.24 cubic feet rice dried and stored. per minute per 100 pounds for 7 days before dry- Dachtler {23) emphasized that the use of the ing, showed a sharp drop in grade. proper airflow rate to dry rice with unheated air INFRARED DRYING OF ROUGH RICE.—^As long ago is of primary concern in bin drying rice. Air as the early 1940's, Texas and California rice- must be supplied at a rate to complete drying growers experimented with industrial infrared before the rice is damaged by mold growth or lights as a heat source for drying rice. By 1948 other causes. Recommendations developed from California workers had reached the conclusion tests in the Southern States and in California are that infrared rice drying was no faster than con- summarized by Dachtler {23). Supplemental heat vection-type driers, and was not as efficient.'^ In is not recommended as a standard practice for bin their six experiments, 7,300 B.t.u.'s were required drying. However, it is desirable to have the neces- to remove 1 pound of water from rice, whereas sary equipment available for use during prolonged convection-type sack rice driers then in use were periods of adverse weather. This normally will using 2,000 B.t.u.'s. include extended periods of high humidity (above Infrared drying of rice has not yet become estab- 75 percent). The temperature of the air entering lished commercially but interest continues. Faulk- the rice may be raised 10° to 15° F. above the am- ner, Wratten, and Miller {S6) obtained highest bient temperature. A maximum of 95° is recom- head rice yields when grain moisture content mended in Texas, and a maximum of 80° to 85° ranged from 21.5 to 23.5 percent. Head rice yields in California. Supplemental heat should be used declined as grain moisture content was decreased until the moisture content of the top foot of rice is from 19.5 to 13.5 percent, especially when drying reduced to 15 percent. After the moisture is re- temperatures exceeded 175° F. Schroeder (118^ duced to this level, unheated air should be used to 119) and Schroeder and Rosberg {120, 121) re- complete the drying to a safe storage level. During ported on laboratory rice-drying experiments. the time unheated air is used, the fan should be Varieties apparently respond differently. The dry- operated only when the relative humidity is less ing rate for the 15-, 20-, and 25-second irradia- than 75 percent, which usually will be during day- tion periods ranged from 1.2 to 1.9 grams of water light hours on clear, bright days. removed per second of irradiation for Nato and AERATING STORED EICE.^—According to Hutch- from 1.3 to 1.6 grams for Magnolia. Tilton and inson and Willms (6'4^), the practice of aerating Schroeder {150) found that rice weevils and lesser grain came into widespread use in the Southwest grain bores in rough rice could be controlled by between 1955 and 1960. They indicate aeration irradiation with infrared lights. They postulated is used (1) to maintain the quality of undried that using gas-fired infrared heating for drying grain until it can be moved through the drier, (2) rough rice would have the additional benefit of to remove harvest or drier heat, (3) to remove killing all stages of these stored-rice insects during small amounts of moisture (1 to 2 percent), and the rice-drying process. In utilizing on-f arm bin drying and storing, the (4) to maintain the quality of grain during stor- rice farmer is assuming considerable responsibil- age. Aeration is defined as the moving of air ity for items that he pays someone else to assume through stored grain at low airflow rates (gen- when he employs commercial facilities. These erally between one-tenth and one-fifth cubic feet include: (1) Considerable supervision over a long per minute per 100 pounds) for purposes other period of time to oversee the drying, particularly than drying, to maintain or improve its value. if he uses unheated air; (2) greater chance of rice Several types of aeration systems are described, deterioration; (3) protection of the rice from as are methods of operation and controls. insect, bird, and animal pests; and (4) protec- Aldred {10) presented the following conclu- sions regarding storage and aeration of rice : ^ UnpubUshed reports from the University of California, (1) Based on information to date (1952), rice Department of Agricultural Engineering. in bulk storage should be turned once every 2 128 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE. months during the winter and at least once a {Sitophilus oryzae (L.) ) attack sound rough rice month during the summer. kernels. The cadelle {Tenebroides mauritanicus (2) With proper aeration, rice with a moisture (L.) ), the saw-toothed grain beetle {Oryzaeohilus content of 18 to 24 percent can be kept for a surinamensis (L.) ), the flat grain beetle {Crypto- week or 10 days without spoilage. lestes pusillus (Schon.) ), the red flour beetle {Tri- (3) For retaining rice with a moisture content holium castaneum, (Herbst)), and the confused of 18 to 24 percent before drying, 2 cubic feet flour beetle {T, confusum (Duv.)) attack broken per minute per 100 pounds is adequate. After the or dehuUed kernels. Moths that infest the sur- rice has been partly dried, one-half cubic foot face of bulk or bagged rough rice and spin web- per minute per 100 pounds is adequate. bing in profusion include the Indian-meal moth (4) Eice with a moisture content below 16 per- {Plodia interpunctella (Hbn.) ), the almond moth cent can be kept several months in either cool or {Ephestia cautella (Wlkr.)), and the rice moth warm weather with the aid of aeration. {Corcyra cephalonica (Staint.)). In storing bran As pointed out by Dachtler {23)^ practically and milled rice, control of the bran bugs, includ- all rough rice is stored in bulk, although sack ing the flour beetles, is important {117^ 152). storage is practical for relatively small lots and for seed rice. A maximum moisture content of Selecfed References 12 percent is recommended for seed rice, but up to 14 percent usually is safe in bulk storage. The (1) ANONYMOUS. 1947. RICE DRYING AND STORAGE IN LOUISIANA. LA. length of the storage period for rice depends on Agr. Expt. Sta. Bui. 416, 22 pp. market conditions, but it usually is from 5 to 8 (2) months. For any common-type storage bin, any 1949. STOP—DON'T PLANT RED RICE. Rice Jour. type of construction material is satisfactory if it 52(2) :: 20. results in a storage structure that will keep the (3) 1962. UNDERGROUND IRRIGATION. Rice Jour. 65(9) : grain dry, cool, and free of insects and other pests 22, 23, illus. and if it provides job safety and convenience (4) ADAIR, C. R. while moving and inspecting the grain. Dachtler 1940. EFFECT OF TIME OF SEEDING ON YIELD, MILL- warned that if dried rough rice is to be stored for ING QUALITY, AND OTHER CHARACTERS IN a few months or longer or if damp rice is to be RICE. Amer. Soc. Agron. Jour. 32: 697-706. (5) and CRALLEY, E. M. held before drying, the storage structure should 1950. 1949 RICE YIELD AND DISEASE CONTROL TESTS. be equipped for aeration. Ark. Agr. Expt. Sta. Rpt. Ser. 15, 20 pp. INSECT INFESTATION OF STORED EICE.—Insect ( 6 ) and ENGLER, KYLE. infestation generally occurs after rice is in stor- 1955. THE IRRIGATION AND CULTURE OF RICE. In age, since most rice harvested and dried by mod- Water, U.S. Dept. Agr. Yearbook of Agr., ern methods is relatively free of insects upon en- pp. 389-394. (7) BEACHELL, H. M., JODON, N. E., and others. tering storage {28), Insect control practices that 1942. COMPARATIVE YIELDS OF TRANSPLANTED AND are used before and during storage have been DIRECT SOWN RICE. Amer. Soc Agron. Jour. developed through research and are widely ap- 34: 129-137. plied. Emphasis now is being placed on the use (8) MILLER, M. D., and BEACHELL, H. M. of protective treatment. Chemicals are now 1962. RICE IMPROVEMENT AND CULTURE IN THE available that can be mixed with or sprayed on UNITED STATES. Adv. in Agron. 14: 61-108, illus. the rice as it is stored and that will protect it (9) ADDICOTT, F. T., and LYNCH, R. S. against invading insects yet leave no residues 1957. DEFOLIATION AND DESICCATION : HARVEST-AID Jiarmful to human beings ( {117^ 152). For infor- PRACTICES. Adv. in Agron. 9: 67-93. (10) ALDRED, F. L. (editor). mation about the use of these chemicals, contact 1953. RECENT RESEARCH ON DRYING AND STORAGE OF your local agricultural authorities. ROUGH RICE. South. Coop. Ser. Bul. 29, 29 Stored rice is subject to attack by a number of pp., illus. insects. The lesser grain borer {Rhyzopertha (11) ANDERSON, K. L., and MCKIE, J. W. dominica (F.)), the Angoumois grain moth 1962. GROWING RICE IN THE MISSISSIPPI DELTA. Miss. Agr. Ext. Serv. Pub. 219, U.S. Dept. {Sitotroga cerealella (Oliv.) ), and the rice weevil Agr. RICE IN THE UNITED STATES 129

(12) ATKINS, J. G., CRALLEY, E. M., and CHILTON, S. J. P. (26) 1957. UNIFOEM RICE SEED TREATMENT TESTS IN 1950. CALIFORNIA RICE PRODUCTION. Calif. Agr. ARKANSAS, LOUISIANA, AND TEXAS, 1955-56. Ext. Serv. Cir. 163, 55 pp., illus. Plant Dis. Rptr. 41: 105-108. (27) DOCKINS, J. O. 1950. PRODUCING QUALITY SEED RICE DEEMED VITAL (13) BAINER, R. 1932. HARVESTING AND DRYING ROUGH RICE IN CALI- TO ARKANSAS RICE INDUSTRY. Rice JOUr. FORNIA. Univ. of Calif., Div. of Agr. Sei. 53(5) :21,35. Bul. 541, 29 pp. illus. (28) EHRLER, W., and BERNSTEIN, L. 1958. EFFECTS OF ROOT TEMPERATURE, MINERAL NU- (14) BAMESBERGER, J. G. TRITION AND SALINITY ON THE GROWTH AND 1954. LAND LEVELING FOR IRRIGATION. U.S. Dept. COMPOSITION OF RICE. Bot. Gaz. 120(2) : Agr. Leaflet 371, 8 pp., illus. 67-74. (15) BARR, H. T., WRATTEN, F. T., POOLE, W. D., and (29) ENGLER, KYLE. WALKER, R. P. 1958. WATER LEVELS IN RICE IRRIGATION WELLS IN 1955. RECOMMENDATIONS FOR BIN DRYING AND THE GRAND PRAIRIE REGION. Ark. Farm Res. STORAGE OF ROUGH RICE IN LOUISIANA. Rev., 7(3) :12. La. Agr. Expt. Sta., Agr. Engin. Dept. Cir. (30) THOMPSON, D. G., and KAZMANN, R. G. 18,16pp. [Processed.] 1945. GROUND WATER SUPPLIES FOR RICE IRRIGATION (16) BLACK, D. E., and WALKER, R. K. IN THE GRAND PRAIRIE REGION, ARKANSAS. 1955. THE VALUE OF PASTURES IN ROTATION WITH Ark. Agr. Expt. Sta. Bui. 457, 56 pp., illus. RICE. La. Agr. Expt. Sta. Bui. 498, 24 pp., (31) EVATT, N. S. illus. 1958. FERTILIZER-WATER DEPTHS TESTS ON RICE, (17) CALDERWOOD, D. L., and HUTCHINSON, R. S. 1956-57. Tex. Agr. Expt. Sta. Prog. Rpt. 1961. DRYING RICE IN HEATED AIR DRYERS WITH 2006, 4 pp. AERATION AS A SUPPLEMENT TREATMENT. (32) and BEACHELL, H. M. U.S. Dept. Agr. Market. Res. Rpt.'508, 22 pp. 1962. SECOND-CROP RICE PRODUCTION IN TEXAS. (18) CHAMBLISS, C. E. Tex. Agr. Prog. 8(6) : 25-28. 1920. PRAIRIE RICE CULTURE IN THE UNITED STATES. (33) and WEIHING, R. M. U.S. Dept. Agr. Farmers' Bui. 1092, 26 pp., 1957. FERTILIZER REQUIREMENTS FOR RICE IN RICE- illus. PASTURE ROTATIONS. Tex. Agr. Expt. Sta. (19) and JENKINS, J. M. Prog. Rpt. 1948, 4 pp. 1925. EXPERIMENTS IN RICE PRODUCTION IN SOUTH- (34) FAULKNER, M. D. WESTERN LOUISIANA. U.S. Dept. Agr. Dept. 1960. WATER PLANTING STUDIES. Rice Jour. 63 Bui. 1356, 32 pp., illus. (7) : 4^50. (20) CHAPMAN, A. L., and PETERSON, M. L. (35) and MiEARs, R. J. 1962. THE SEEDLING ESTABLISHMENT OF RICE UNDER 1962. LEVELING RICE LAND AND WATER. La. Agr. WATER IN RELATION TO TEMPERATURE AND 5(4) : 3,16, illus. DISSOLVED OXYGEN. Crop Sci. 2(5) I 391- (36) WRATTEN, F. T., and MILLER, M. F. 395. 1968. INFRARED ENERGY FOR PROCESSING ROUGH (21) CLARK, F., NEARPASS, D. C, and SPECHT, A. W. RICE. 12tli Rice Tech. Working Group 1957. INFLUENCE OF ORGANIC ADDITIONS AND FLOOD- Proc, March 5-7, New Orleans, La. p. 58. ING ON IRON AND MANGANESE UPTAKE BY RICE. (37) FiNFROCK, D. C, and MILLER, M. D. Agron. Jour. 49 : 586-589. 1958. ESTABLISHING A RICE STAND. Uuiv. Of Calif., (22) CuRLEY, R. G., and Goss, J. R. Div. of Agr. Sei. Leaflet 99, 12 pp., illus. 1964. ESTIMATING COMBINE LOSSES IN RICE. Uuiv. (38) RANEY, F. M., MILLER, M. D., and BOOHER, Calif., Dept. Agr. Engin., 4 pp. [Unnum- L.J. bered. Mimeographed.] 1960. WATER MANAGEMENT IN RICE PRODUCTION. Univ. Of Calif. Div. of Agr. Sei. Leaflet 131, (23) DACHTLER,W. C. (editor). 2 pp., illus. 1959. RESEARCH AND CONDITIONING AND STORAGE OF (39) -ViSTE, K. L., HARVEY, W. A., and MILLER, ROUGH AND MILLED RICE A REVIEW THROUGH M. D. 1958. U.S. Dept. Agr., Agr. Res. Serv. ARS 1957. WEED CONTROL IN RICE. Uuiv. of Calif. Div. 20-7, 55 pp., illus. of Agr. Sei. Leaflet 97, 2 pp., illus. (24) DAVIS, J. H., SONNIER, E. A., and WHITE, T. W. (40) GARRISON, R. H. 1963. PASTURE-RICE ROTATIONS FOR SOUTHWEST 1959. GOOD SEED DOES NOT COST, IT PAYS HANDSOME LOUISIANA. La. Agr. 7(1) : 4-5. DIVIDENDS. Seedmen's Digest, Aug., pp. 22, (25) DAVIS, L. L. 23, 44, illus. 1944. HARVESTING RICE FOR MAXIMUM MILLING (41) GATTIS, J. L., KOCH, K. A., and MCVEY, J. L. QUALITY IN CALIFORNIA. Ricc Jour. 47 (3) : 1959. LAND GRADING FOR SURFACE IRRIGATION. Ark. 3-4,17-18, illus. Agr. Ext. Serv. Cir. 491, 29 pp., illus. 130 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE'

(42) GAY, B. (58) HILDRETH, R. J., and SORENSON, J. W., JR. 1961. 'RATOONING' METHODS OF A TEXAS RICE 1957. PROFITS AND LOSS FROM ON-FARM DRYING AND FARMER. Rice Jour. 64(2) : 22. STORAGE OF RICE IN TEXAS. TeX. Agr. Expt. (43) GERLOW, A., and MULLINS, T. Sta. Bui. 865, 16 pp. 1958. RESERVOIRS FOR IRRIGATION IN THE GRAND (59) HiNKLE, D. A. PRAIRIE AREA I AN ECONOMIC APPRAISAL. 1952. FIELD DRYING OF RICE BY CHEMICALS. SOUth. Ark. Agr. Expt. Sta. BuL 606, 24 pp. Weed Conf. Proc. 5: 175-177. (44) Goss, W. L., and BROWN, E. (60) 1939. BURIED BED RICE SEED. Agron. JouF. 31: 633- 1953. EFFECT OF PRB-HARVEST CHEMICALS UPON 637. COMBINE EFFICIENCY AND MILLING YIELD OF (45) GRAY, L. C, and THOMPSON, E. K. RICE. South. Weed Conf. Proc. 6: 72-75. 1941. HISTORY OF AGRICULTURE IN THE SOUTHERN (61) U.S. TO 1860. Carnegie Inst. Wash. Pub. 1954. PRE-HARVEST TREATMENT OF RICE AS AN AID 430. Peter Smith, New York. [Reprinted IN DRYING. Rice Tech. Working Group in 2 vols.] Proc. 6: 16-17. (46) GREEN, B. L. (62) HuEY, B. A., and CHAPMAN, S. L. 1961. FISH FARMING ; PAST, PRESENT, FUTURE. ARK. 1970. NITROGEN FERTILIZATION OF RICE IN ARKAN- Agr. Econ. 3(4) : 1,2. SAS. Ark. Agr. Ext. Serv. Leaflet 405 (47) and MULLINS, T. (Rev.), illus. 1959. USE OF RESERVOIRS FOR PRODUCTION OF FISH (63) HURST, W. M., and HUMPHRIES, W. R. IN THE RICE AREAS OF ARKANSAS. Ark. Agr. 1955. HARVESTING WITH COMBINES. U.S. Dept. Expt. Sta. Spec. Rpt. 9, 16 pp. Agr. Farmers' Bui. 1761, 40 pp., illus. (48) and WHITE, J. H. (64) HuTCHiNSON, R. S., and WILLMS, E. F. 1963. COMPARISON OF THREE SELECTED ROTATIONS 1962. OPERATING GRAIN AERATION SYSTEMS IN THE IN EASTERN ARKANSAS. Ark. Agr. Expt. Sta. SOUTHWEST. U.S. Dept. Agr. Market. Res. Bui. 664, 20 pp., illus. Rpt. 512, 20 pp., illus. (49) HALICK, J. V. (65) JENKINS, J. M., and JONES, J. W. 1960. EFFECT OF TEMPERATURE DURING RIPENING ON 1944. RESULTS OF EXPERIMENTS WITH RICE IN LOUI- QUALITY CHARACTERISTICS OF RICE. Rice SIANA. La. Agr. Expt. Sta. Bui. 384, 39 pp. Tech. Working Group Proc, June 29 to (66) JODON, N. E. July 1, Lafayette, La. MP 488, p. 14. 1953. GROWING PERIOD OF LEADING RICE VARIETIES (50) HALL, V. L. 1959. GREENHOUSE STUDIES ON THE RELATION OF WHEN SOWN ON DIFFERENT DATES. La. Agr. WATER MANAGEMENT TO THE GROWTH OF RICE. Expt. Sta. Bui. 476, 8 pp. Ark. Agr. Expt. Sta. Rpt. Ser. 89, 22 pp. (67) JOHNSTON, T. H., ADAIR, C. R., TEMPLETON, G. E., (51) HALL, V. L. and others. 1960. WATER SEEDING OF RICE IN ARKANSAS. Rice 1963. NOVA AND VEGOLD—NEW RICE VARIETIES. Jour. 63(13) : 13. Ark. Agr. Expt. Sta. Bui. 675, 24 pp., illus. (52) SIMS, J. L., and JOHNSTON, T. H. (68) and CBALLEY, E. M. 1968. TIMING OF NITROGEN FERTILIZATION OF RICE. 1955. RICE VARIETIES AND THEIR YIELDS IN ARKAN- II. CULM ELONGATION AS A GUIDE TO OPTIMUM SAS, 1948^1954. Ark. Agr. Expt. Sta. Rpt. TIMING OF APPLICATIONS NEAR MIDSEASON. Ser. 49, 20 pp. Agron. Jour. 60: 450-453. (69) CRALLEY, E. M., and HENRY, S. E. (53) and THOMPSON, L. F. 1959. PERFORMANCE OF RICE VARIETIES IN ARKAN- 1962. SALINITY AND ALKALINITY OF RICE SOILS IN SAS, 1953-1958. Ark. Agr. Expt. Sta. Rpt. ARKANSAS. Ark. Farm Res. 11(2) : 11. Ser. 85, 31 pp., illus. (54) HASKELL, C. G. (70) TEMPLETON, G. E., SIMS, J. L., and others. 1915. IRRIGATION PRACTICE IN RICE GROWING. U.S. 1966. PERFORMANCE IN ARKANSAS OF NOVA 66 AND Dept. Agr. Farmers' Bui. 673, 12 pp. OTHER MEDIUM-GRAIN RICE VARIETIES, 1960 TO (55) HENDERSON, S. M. 1965. Ark. Agr. Expt. Sta. Rpt. Ser. 148, 1954. THE CAUSES AND CHARACTERISTICS OF RICE 24 pp. CHECKING. Rice Jour. 57(5) : 16, 18. (56) (71) JONES, J. W., ADAIR, C. R., JODON, N. E., and others. 1955. DEEP-BED RICE DRIER PERFORMANCE. Agr. En- 1947. EFFECT OF ENVIRONMENT AND SOURCE OF SEED gin. 36: 817-820. ON YIELD AND OTHER CHARACTERS IN RICE, (57) Amer. Soc Agron. Jour. 39: 874-886. 1958. DEEP-BED GRAIN DRYING ON THE RANCH WITH (72) DAVIS, L. L., and WILLIAMS, A. H. UNHEATED AIR. Uuiv. of Calif. Div. of Agr. 1950. RICE CULTURE IN CALIFORNIA. U.S. Dept. Sei. Leaflet 103, 2 pp. Agr. Farmers' Bui. 2022, 32 pp., illus. RICE IN THE UNITED STATES 131

(73) DocKiNS, J. O., WALKER, R. L., and DAVIS, (86) MARR, J. C. W.C. 1957. GRADING LAND FOR SURFACE IRRIGATION. 1952. RICE PRODUCTION IN THE SOUTHERN STATES. Univ. of Calif. Div., Agr. Sei. Cir. 438, 48 U.S. Dept. Agr. Farmers' Bul. 2043, 36 pp, pp., illus. illus. (87) MiKKELSEN, D. S., FiNFROCK, D. C, and MILLER, M. D. (74) JENKINS, J. M., WYCHE, R. H., and NEL- 1958. RICE FERTILIZATION. Calif. Agr. Expt. Sta. SON, M. 1938. RICE CULTURE IN THE UNITED STATEb. Ext. Serv. Leaflet 96, 12 pp., illus. U.S. Dept. Agr. Farmers' Bul. 1808, 29 pp. (88) and GLAZEWSKI, A. T. 1963. OCCURRENCE AND PHYSIOLOGICAL NATURE OF illus. ENDOGENOUS GROWTH SUBSTANCES IN HULLS (75) KAPP, L. C. OF oRYZA SATIVA. Rice Tech. Work. Group 1947. THE EFFECT OF COMMON SALT ON RICE PRO- Proc. 1962. Ark. Agr. Exp. Sta. Unnumbered DUCTION. Ark. Agr. Expt. Sta. Bui. 465, pub. pp. 24-25. 7 pp. (89) MORRISON, S. R. RESTER, E. B. (76) 1953. RICE IRRIGATION TESTS AT THE BEAUMONT 1959. EFFECTS OF CERTAIN PREPROCESSING AND CUL- STATION, 1952. Tex. Agr. Expt. Sta. Prog. TURAL VARIABLES UPON MILLING AND OTHER Rpt. 1542, 2 pp. PROCESSING QUALITIES OF RICE. Calif. Rlce (90) DAVIS, W. C, and SORENSON, J. W., JR. Res. Symp. Proc, Albany, Calif., pp. 6-12. 1954. BIN DRYING OF RICE AT THE RICE-PASTURE (77) and PENCE, J. W. EXPERIMENT STATION, 1953-54. Tex. Agr. 1962. RICE INVESTIGATIONS AT WESTERN REGIONAL Expt. Sta. Prog. Rpt. 1670, 9 pp. RESEARCH LABORATORY. Rice Jour. 65(7) : (91) MORSE, M. D., LINDT, J. H., OELKE, E. A., and other«. 45-47, illus. 1967. THE EFFECT OF GRAIN MOISTURE AT TIME OF (78) KING, B. M. HARVEST ON YIELD AND MILLING QUALITY OF 1937. THE UTILIZATION OF WABASH CLAY (GUMBO) RICE. Rice Jour. 70(11) : 16-20. SOILS IN CROP PRODUCTION. Mo. Agr. Expt. (92) MucKEL, D. C. Sta. Bui. 254, pp. 12-32. 1959. REPLENISHING UNDERGROUND WATER SUPPLIES (79) LouRENCE, F. J., PRUITT, W. O., and SERVIS, ALLEN. ON THE FARM. U.S. Dept. Agr. Leaflet 452, 1970. ENERGY BALANCE AND THE CROP WATER RE- 8 pp., illus. QUIREMENTS OF RICE GROWN IN CALIFORNIA. (93) MuLLiNS, T. Dept. Water Sei. and Engin., Univ. of Cali- 1954. ECONOMIC APPRAISAL OF FARMING PRACTICES fornia, Davis, Water Sei. and Engr. Paper AND ROTATION PROGRAMS OF LOUISIANA RICE FARMS. La. Agr. Expt. Sta. Bui. 491, 40 9002, 44 pp. pp. (80) MCFARLANE, V. H., HOGAN, J. T., and MCLEMORE, (94) T. A. 1960. PRODUCTION PRACTICES, COSTS AND RETURNS 1955. EFFECTS OF HEAT TREATMENT ON THE VIA- FOR MAJOR ENTERPRISES ON RICE FARMS IN BILITY OF RICE. U.S. Dept. Agr. Tech. Bui. THE DELTA AREA OF MISSISSIPPI. MÍSS. Agr. 1129, 51 pp., illus. Expt. Sta. Bui. 595, 24 pp. (81) MACKIE,W. W. (95) and SLUSHER, M. W. 1943. RICE IN THE IMPERIAL VALLEY (CALIFORNIA). 1950. COMPARISON OF FARMING SYSTEMS FOR SMALL Imper. Rice Growers' Coop. Assoe. [Un- RICE FARMS IN ARKANSAS. Ark. Agr. Expt. numbered Rpt.] Sta. Bui. 498, 44 pp., illus. (96) and SLUSHER, M. W. (82) MCNEAL, XziN. 1951. COMPARISON OF FARMING SYSTEMS FOR LARGE 1949. ARTIFICIAL DRYING OF COMBINED RICE. Ark. RICE FARMS IN ARKANSAS. Ark. Agr. Expt. Agr. Expt. Sta. Bui. 487, 30 pp. Sta. Bul. 509, 40 pp. (83) (97) NELSON, G. S. 1950. EFFECT OF COMBINE ADJUSTMENT ON HARVEST 1961. AERIAL APPLICATIONS OF GRANULAR AND PEL- LOSSES OF RICE. Ark. Agr. Expt. Sta. Bui. LETED MATERIALS AND SEEDS. Rice Jour. 500, 26 pp., illus. 64(7) : 2^27. (84) (98) NELSON, M. 1950. WHEN TO HARVEST RICE FOR BEST MILLING 1931. PRELIMINARY REPORT ON CULTURAL AND FER- QUALITY AND GERMINATION. Ark. Agr. Expt. TILIZER EXPERIMENTS WITH RICE IN ARKANSAS. Sta. Bui. 504, 41 pp. Ark. Agr. Expt. Sta. Bui. 264, 46 pp. (85) (99) 1961. EFFECTS OF DRYING TECHNIQUES AND TEM- 1944. ROTATION, CULTURAL AND IRRIGATION PRAC- PERATURES ON HEAD RICE YIELDS. Ark. Agr. TICES AFFECTING RICE PRODUCTION. Ark. Expt. Sta. Bui. 640, 22 pp., illus. Agr. Expt. Sta. Bui. 445, 45 pp. 132 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

(100) OELKE, E. A., BALL, R. B., WICK, C. M., and MILLER, (116) and ROUSE, PHIL. M. D. 1965. THE BIOLOGY AND ECOLOGY OF THE GRAPE 1969. INFLUENCE OF GRAIN MOISTURE AT HARVEST ON COLASPIS, COLASPIS FLA VIDA SAY, IN RELATION SEED YIELD, QUALITY, AND SEEDLING VIGOR OF TO RICE PRODUCTION IN THE ARKANSAS GRAND RICE. Crop Sei. 9: 144-147. PRAIRIE. Ark. Agr. Expt. Sta. Bui. 694, (101) MORSE, M. D., and MIKKELSEN, D. S. 31pp. 1967. RICE STAND ESTABLISHMENT. Univ. Of Calif., (117) ROUSE, PHIL, ROLSTON, L. H., and LINCOLN, C. Div. of Agr. Sei. Leaflet 196, 10 pp., illus. 1968. INSECTS IN FARM-STORED RICE. Ark Agr. (102) and MUELLER, K. E. Expt. Sta. Bui. 600, 25 pp. 1969. INFLUENCES OF WATER MANAGEMENT AND FER- (118) SCHROEDER, H. W. TILITY ON RICE GROWTH AND YIELD. AgrOn. 1960. INFRA-RED DRYING OF ROUGH RICE. II. Jour. 61: 227-230. SHORT GRAIN TYPE CALROSE AND CALORO. (103) OvERSTREET, R., and SCHULZ, R. K. Rice Jour. 63(13) : 6-8, 25-28. 1958. THE EFFECT OF RICE CULTURE ON A NON- (110) SALINE SODIC SOIL OF THE FRESNO SERIES. 1961. INFRA-RED DRYING OF ROUGH RICE. IH. Hilgardia 27(12) : 319-^32, illus. MEDIUM-GRAIN TYPE NATO AND MAGNOLIA. (104) PEARSON, G. A. Rice Jour. 64(1) : 11-12, 24-27, illus. 1959. FACTORS INFLUENCING SALINITY OF SUB- (120) • and ROSBERG, D. W. MERGED SOILS AND GROWTH OF CALORO RICE, 1959. DRYING ROUGH-RICE WITH INFRA-RED RADIA- Soil Sei. 87(4) : 198-206. TION. Tex. Agr. Expt. Sta. MP-354 (May), (105) 4 pp. 1961. THE SALT TOLERANCE OF RICE. Intematl. ( 121 ) and ROSBERG, D. W. Riee Oomm. Newsletter 10(1) : 1^. 1960. INFRA-RED DRYING OF ROUGH RICE. I. LONG- (106) and AYERS, A. D. GRAIN TYPE—^REXORO AND BLUEBONNET 50. 1960. RICE AS A CROP FOR SALT-AFFECTED SOIL IN Rice Jour. 63(12) : 3-5, 23-27, illus. PROCESS OF RECLAMATION. U.S. Dept. Agr. (122) SCOTT, V. H., LEWIS, D. C, FOX, D. R., and BABB, Prod. Res. Rept. 43, 13 pp., illus. A. F. (107) and BERNSTEIN, L. 1961. PLASTIC LEVEES FOR RICE IRRIGATION. Calif. 1959. SALINITY EFFECTS AT SEVERAL GROWTH Agr. 15(11) :8, 9. STAGES OF RICE. Agpou. JouT. 51:654-657. (123) SENEWIRATNE, S. T., and MIKKELSEN, D. S. (108) PERKINS, W. R., and LUND, C. F. 1961. PHYSIOLOGICAL FACTORS LIMITING GROWTH 1950. CROP ROTATION ON RICE FARMS. Ark. Agr. AND YIELDS OF ORYZA SATIVA UNDER UN- Ext. Serv. Leaflet 134. FLOODED CONDITIONS. Plant and Soil 14: (109) PoMiNSKi, J., WASSERMAN, T., SPADARO, J. J., and 127-146. others. (124) SIMMONS, C. F. 1961. IMPROVEMENTS IN COMMERCIAL DRYING OF 1940. RICE PRODUCTION AND RICELAND USES IN SOUTHERN GROWN RICE. I. ZENITH—A ARKANSAS. Ark Agr. Ext. Serv. Cîr. 424, MEDIUM-GRAIN VARIETY. Rice Jour. 64(9) : 16 pp. 10, 12-13, 16-17. (125) SIMS, J. L. (110) RANEY, F. C. 1961. FISH-RICE PROJECT PLANS. RicC. JoUr. 1959. WARMING BASINS AND WATER TEMPERATURE. 64(7) :22. Calif. Rice Res. Symp. Proc., Albany, Calif., (126) pp. 20-23. 1964. NITROGEN FERTILITY STUDIES IN RICE FIELDS (111) HAGAN, R. M., and FINFROCK, D. C. AND RESERVOIRS. RíCC Jour. 67(7) : 11, 13. 1957. WATER TEMPERATURE IN IRRIGATION. Calif. (127) Agr. 11(4) : 19, 20, 37, illus. 1964. NITROGEN AVAILABILITY IN RICE FIELD AND (112) REED, J. F., and STURGIS, M. B. RESERVOIR SOILS. Ark. Farm Res. 13(2) : 2. 1936. TOXICITY FROM ARSENIC COMPOUNDS TO RICE (128) ON FLOODED SOILS. Amer. Soc. Agron. Jour. 1964. GROWTH OF ARABLE CROPS AS A MEANS OF 28: 432-436. REDUCING AVAILABLE NITROGEN IN RESERVOIR (113) REYNOLDS, E. B. SOIL. Ark. Farm. Res. 13(3) : 5. 1954. RESEARCH ON RICE PRODUCTION IN TEXAS. (129) HALL, V. L., and JOHNSTON, T. H. Tex. Agr. Expt. Sta. Bui. 775, 29 pp., illus. 1967. TIMING OF N FERTILIZATION OF RICE. I. EF- (114) ROBERTSON, R. D. FECT OF APPLICATIONS NEAR MIDSEASON ON 1917. IRRIGATION OF RICE IN CALIFORNIA. Calif. VARIETAL PERFORMANCE. AgrOU. JOUr. 59 I Agr. Expt. Sta. Bui. 279, pp. 253-270, illus. 63^66. (115) ROLSTON, L. H., and ROUSE, PHIL. (130) SLUSHER, M. W. 1960. CONTROL OF GRAPE COLASPIS AND RICE WATER 1953. THE USE OF AIRPLANES ON RICE FARMS IN WEEVIL BY SEED OR SOIL TREATMENT. Ark. ARKANSAS. Ark. Agr. Expt. Sta. Bui. 541, Agr. Expt. Sta. Bui. 624, 10 pp. 20 pp. RICE IN THE UNITED STATEß 133

(146) STOUT, B. A. (131) 1955. ENTERPRISE COSTS AND RETURNS ON RICE 1966. EQUIPMENT FOR RICE PRODUCTION. FAO FARMS. Ark. Agr. Expt. Sta. Bul. 549, 34 Devlpmt. Paper 84, 169 pp., illus. United pp., illus. Nations, Rome. (132) and MuLijNS, T. (147) STROMBERG, L. K., and YAMADA, H. 1948. MECHANIZATION OF THE RICE HARVEST. Ark. 1955. WATER QUALITY IN RICE FIELDS. Galif. Agr. Agr. Expt. Sta. Rept. Ser. 11, 32 pp., illus. 9(3) :10. (133) -and MuLLiNS, T. (148) SULLIVAN, K. B. 1952. RICE MILL YIELD AND GRADE IN RELATION TO 1960. WATER IS A **CROP" WITH ROTATION RESER- VARIETY AND METHOD OF HARVEST. Ark. VOIRS. Ark. Farmer 62(8) : 4, illus.

Agr. Expt. Sta. Bui. 526, 36 pp., illus. (149) THOMPSON, W. R., and WALLER, T. M. (134) SMITH, R. J., JR., HINKLE, D. A., and WILLIAMS, 1952. GROWING RICE IN THE MISSISSIPPI DELTA. F. J. Miss. Agr. Ext. Serv. Pub. 219, 6 pp. 1959. PRE-HARVEST DESICCATION OF RICE WITH (150) TiLTON, E. W., and SCHROEDER, H. W. CHEMICALS. Ark. Agr. Expt. Sta. Bui. 619, 1961. THE EFFECT OF INFRARED RADIATION ON IM- 16 pp., illus. MATURE INSECTS IN KERNELS OF ROUGH RICE. (135) SMITH, W. D. Rice Jour. 64(9) : 23-25. 1940. HANDLING ROUGH RICE TO PRODUCE HIGH (151) TULLíS, E.G. GRADES. U.S. Dept. of Agr. Farmers' Bui. 1951. HERBICIDES FOR ACCELERATING MATURATION 1420, 21 pp., illus. OF RICE. South. Weed Gonf. Proc. 4: 1-2. (136) -DEFFES, J. J., BENNETT, 0. H., and HURST, (152) UNITED STATES DEPARTMENT OF AGRICULTURE. W. M. 1957. CONTROLLING INSECT PESTS OF STORED RICE. 1930. DRYING COMBINE HARVESTED RICE ON THE U.S. Dept. Agr. Agr. Handb. 129, 30 pp.. FARM. U.S. Dept. Agr., U.S. Grain Stand- illus. ards Act-Grain Investigations 57, 20 pp. (153) WALKER, R. K. and STURGIS, M. B. (137) DEFFES, J. J., BENNETT, C. H., and others. 1946. A TWELVE-MONTH GRAZING PROGRAM FOR THE 1933. ARTIFICIAL DRYING OF RICE ON THE FARM. RICE AREA OF LOUISIANA. La. Agr. Expt. U.S. Dept. Agr. Gir. 292, 24 pp., illus. Sta. Bui. 407,19 pp. (138) DEFFES, J. J., BENNETT, C. H., and others. 1938. EFFECT OF DATE OF HARVEST ON YIELD AND (154) WASSERMAN, T., FERREL, R. E., BROWN, A. H., and MILLING QUALITY OF RICE. U.S. Dept. Agr. SMITH, G. S. Gir. 484. 20 pp. 1957. COMMERCIAL DRYING OF WESTERN RICE. Gereal (139) SON NIER, A. Ghem. Today 2(9) : 251-254. 1960. CATFISH, CRAYFISH AND RICE. Rice. Jour. (155) FERREL, R. E., KAUFMAN, V. F., and others. 63(5) : 6, 8, 9, illus. 1958. IMPROVEMENTS IN COMMERCIAL DRYING OF (140) SoRENSoN, J. W., JR. WESTERN RICE. I. MIXING TYPE DRYER— 1957. SUPPLEMENTAL HEAT FOR DRYING RICE IN FARM LOUISIANA STATE UNIVERSITY DESIGN. Rice STORAGE BINS. Tex. Agr. Expt. Sta., Dept. Jour. 61(4) : 30-32, 34-36, 38; 61(5) : 40, Agr. Engin., 9 pp. [Processed Rpt.] 42-46. (141) (156) FERREL, R. E., KAUFMAN, Y. F., and others. 1958. RECOMMENDATIONS FOR DRYING AND STORING 1958. IMPROVEMENTS IN COMMERCIAL DRYING OF RICE IN FARM STORAGE BINS. TeX. Agr. Expt. WESTERN RICE. II. NON-MIXING COLUMNAR- Sta., Dept. Agr. Engin. [Processed Rpt.] TYPE DRYER. Rice Jour. 61(7) : 9-12, 14. (142) and DAVIS, W. G. ( 157) WASSON, R. A., and WALKER, R. K. 1955. DRYING AND STORING ROUGH RICE IN FARM 1955. LOUISIANA RICE. La. Agr. Ext. Serv. Ext. STORAGE BINS, 1954-55. Tex. Agr. Expt. Sta. Pub. 1182,16 pp., illus.

Prog. Rpt. 1819, 8 pp. (158) WEBSTER, R. K., HALL, D. H., HEERES, JACOB, and (143) DAVIS W. G., and HOLLINGSWORTH, J. P. others. 1948. DRYING RICE IN SACKS. Tex. Agr. Expt. Sta. 1970. ACHLYA KLEBSIANA AND PYTHIUM SP. AS PRI- Prog. Rpt. 1138, 4 pp. MARY CAUSES OF SEED ROT AND SEEDLING DIS- (144) SPADARO, J. J. EASE OF RICE IN CALIFORNIA. Phytopathol- 1961. EVALUATION OF WRRL DRYING METHODS ON ogy 60: 964-968. SOUTHERN RICE. Sccond Gonf. on Rice Utili- (159) WEIHING, R. M., MONCRIEF, J. B., and DAVIS, W. G. zation Proc, May 18-19. U.S. Dept. Agr., 1950. YEARLONG GRAZING IN THE RICE-PASTURE SYS- Agr. Res. Serv. ARS 74-24, pp. 22-23. TEM OF FARMING. Tcx. Agr. Expt. Sta. (145) STANSEL, J. W., HALICK, J. V., and KRAMER, H. H. Prog. Rpt. 1280, 4 pp. 1961. INFLUENCE OF TEMPERATURE ON HEADING WILLIAMS, DATES AND GRAIN CHARACTERISTICS OF RICE. (160) A. H. Rice Tech. Working Group Proc, June 29 1952. PRE-HARVEST DRYING OF RICE BY CHEMICAL to July 1, Lafayette, La. MP 488, pp. 14-15. TREATMENT. Dowu to Earth 8(2) : 2-3. 134 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

^^^^^ Tn^o'^"""' ^- ^" ^""^ FiNFRocK, D. C. (163) MiKKELSEN, D. S., MuELLER, K. E., and RUCK- 1962. EFFECT OF PLACEMENT AND TIME OF INCOR- MAN, J. E. PORATION OF VETCH ON RICE YIELDS. AgrOn. 1968. NITROGEN IMMOBILIZATION BY RICE STRAW /-.^ox ^^^^' 54(6) : 547-549. INCORPORATED IN LOWLAND RICE PRODUCTION. (162) FiNFRocK, D. C, and MILLER, M. D. Plant and Soil 28(1) : 111-122 1957. GREEN MANURES AND CROP RESIDUES IN (164) WYCHE, R. H., and CHEANEY, R. L. MANAGING RICE SOILS. Calif. Agr. Expt. 1955. WATER SEEDING OF RICE. Tex. Agr Expt Sta Sta. Ext. Serv. Leaflet 90, 6 pp. Prog. Rpt. 1778, 3 pp. WEEDS AND THEIR CONTROL

R. J. SMITH, JR., and D. E. SEAMAN

Losses Due to Weed^ and Cost of Weed redstem {Ammannia auriculata Willd.), and wa- Control terhyssop {Bacopa rotundifolia (Michx.) Wettst.). Emersed aquatic weeds of California Weeds compete with rice {Oryza sativa L.) for ricefields include burhead {Echinodorus cordi- light, nutrients, space, water, and other growth re- folius (L.) Griseb.), California arrowhead {Sag- quirements. Weeds reduce grain yields, lower the ittaria calycina Engelm.), ducksalad, narrowleaf market value of the crop by reducing quality, and cattail {Typha angustifolia L.), ammannia, red- increase the cost of production, harvesting, drying, stem, waterhyssop, and common waterplantain and cleaning. The use of herbicides by rice farmers {Alisma triviale Pursh). in the United States has steadily increased. About Submersed weeds common in water-seeded rice 80 to 90 percent of the commercial rice in the in California include American pondweed {Po- United States is treated each year with one or more tamogetón nodosiis Poir.), horned pondweed herbicides. {Zannichellia palustris L.), and naiad {Najas Problem Weeds spp.) {6). Sedge weed that infest rice in the Southern Problem weeds of ricefields are grouped as weed States include beakrush {Rhynchospora cornicu- grasses; broadleaf, aquatic, and sedge weeds; and lata (Lam.) Gray), bulrush {Scirpus spp.), fim- algae (4, 16), bristylis {Fimbristylis spp.), spikerush {Eleo- Weed grasses that infest rice in the Southern charis spp.), and umbrellaplant {Cyperus spp.). States include barnyardgrass {Echinochloa spp.), Sedge weeds of California ricefields include rough- broadleaf signalgrass {Brachiaria platyphylla seed bulrush {Scirpus mucronatus L.), large- (Griseb.) Nash), red rice {Oryza sativa L.), and spiked spikerush {Eleocharis macrostachya sprangletop {Leptochloa spp.). Weed grasses of Britt.), and smallflower umbrellaplant {Cyperus California ricefields include barnyardgrass and difformisLi.), bearded sprangletop {Leptochloa fascicularis Several algae infest ricefields. One of these is (Lam.) Gray). chara {Chara spp.). Other algae include green Broadleaf weeds that infest rice in the Southern algae {Chlorophyceae), such as Hydrodictyon States include dayflower {Oommelina communis spp., Pithophora spp., and Spirogyra spp., and L.), eclipta {Eclipta alba (L.) Hassk.), hemp ses- blue-green algae {Cyanophyceae)^mQ\.\Jidimg Ana- bania (Seslania exaltata (Eaf.) Cory), Indian haena spp., Lynghya spp., Nostoc spp., and Phorm- joint vetch {Aeschynomene indica L.), mexican- idiiim spp. All these algae frequently form scum weed {Oaperonia castaneaefolia (L.) St. Hil.), in ricefields. morningglory {Ipomoea spp.), northern joint- vetch {Aeschynomene virginica (L.) B.S.P.), and Losses From Competition of Specific Weeds smartweed {Polygonum spp.). None of these weeds is troublesome in California. The competitive effects of a particular weed are Emersed aquatic weeds that infest rice in the influenced by many factors. The most important Southern States include alligatorweed (Alteman- are: Soil fertility or fertilizer applications; the thera philoxeroides (Mart.) Griseb.), ammannia population, vigor, height, and growth duration (Ammannia spp.), ducksalad {Heteranthera lim- of the weed; and the stand, variety, vigor, and osa (Sw.) Willd.), false pimpernel {Lindemia growth stage of the rice during the period of spp.). gooseweed {Sphenoclea zeylanica Gaertn.), competition. mudplantain {Heteranthera reniformis E. & P.) Some weed species reduce rice yields more than 135 136 AGRICULTURE HANDBOOK NO. 289, U.S. DSPT. OF AGRICULTURE

others. Heavy infestations of barnyardgrass (five In a 10-year experiment in Arkansas, cropping plants per square foot) and northern joint vetch systems included either soybeans {Glycine max (one plant per square foot), competing all season (L.) Merr.) or rice growing continuously, or ro- with optimum stands of rice, reduced yields 49 tations of soybeans and rice in a 2-year cycle {15). and 19 percent, respectively {12). All plots contained low weed infestations in 1960 Short weeds compete less with rice than tall when the experiment began. In plots planted con- weeds. All-season competition of ducksalad, a tinuously to rice, infestations of barnyardgrass, short aquatic weed, reduced yields 21 percent; but red rice, and umbrellaplant increased substan- all-season competition of barnyardgrass and hemp tially; but in plots planted alternately with rice sesbania, which grow taller than rice, reduced and soybeans, these weeds increased only slightly. yields 49 and 40 percent, respectively {12), In California, rotations of rice with fallow or A good stand of rice can withstand weed com- with an unirrigated row crop such as safñower petition better than a thin stand. In a rice stand of {Carthamus tinctorim L.) control cattail, bul- only three plants per square foot, one barnyard- rush, spikerush, American pondweed, and other grass plant per square foot reduced the yield 57 perennial weeds that form large root or rhizome percent; but one barnyardgrass plant per square systems, tubers, or winter buds in the soil. Deep foot in a rice stand of 31 plants per square foot plowing to break up and dry out propagation parts reduced the yield only 25 percent, which was only of these perennials during the fallow season im- 5 percent more than the difference in yield between proves control. the two rice stands under weed-free competition Rice following poorly drained native pastures conditions {12), is usually heavily infested with perennial weeds such as spikerush and jointed flatsedge {Cyperus Weed Control Practices articidatus L.). Rice following drained and heav- Effective weed control systems combine preven- ily fertilized improved pastures usually is heavily tive, mechanical, cultural, and chemical methods. infested with barnyardgrass but not with peren- Nonchemical methods may combine some or all of nial weeds. the following practices: Planting weed-free rice Red rice is usually spread by planting contami- seed, crop rotation, leveling land, seedbed prepa- nated seed {16)^ and the continued wide use of ration, selecting the proper seeding method, and clean certified seed has nearly eliminated this managing water and fertilizers properly. Chemi- problem in California. Because herbicides do not cal methods involve the use of herbicides that control red rice selectively in the rice crop, scat- selectively control weeds in rice when applied tered infestations of this and other weeds should correctly. be removed by hand from seed-rice fields to avoid lowered quality and to prevent further spread of Preventive, Mechanical, and Cultural Control the weeds. Practices that help prevent weed infestations Plowing, disking, harrowing, rotary tilling, or or their spread in clean fields include the use of combinations of these mechanical methods are high-quality rice seed that is free of weed seed, ir- used to prepare ricefield seedbeds and to eliminate rigation with water free of weed seeds or other young weeds {16), One of the main goals of all weed propagules, and cultivation with clean methods of seedbed preparation is to eliminate all equipment. High-quality seed also produce vigor- weed growth up to the time of seeding. The meth- ous seedlings that compete more effectively with od chosen depends on the soil type and condition, weeds. other crops in the rotation, method of seeding, Eotary hoeing soon after crop emergence con- climate, and the kinds of weeds present. Young trols small weeds in dry-seeded rice {16), It is weeds that emerge after seeding are more easily the only practical method of cultivation after seed- controlled by herbicides than old weeds that may ing and is most effective when the soil is neither survive an incomplete seedbed preparation, so too dry nor too wet. thoroughness of preparation is important. The occurrence of a particular weed species in Herbicides used in rice, and in crops rotated rice is often associated with the crop rotation used. with rice, control weeds better than crop rotation RICE IN THE UNITED STATES 137

alone. Herbicides applied before or after emer- Draining the field as soon as green areas of algae gence of row and pasture crops reduce weed infes- appear on the soil and then alternately flooding tations in the rice crop. Likewise, herbicides used and draining until the rice is tall enough to to control weed grasses and other weeds in the rice shade the soil and water also control or reduce crop reduce weed infestations in crops rotated algal growths (4). However, barnyardgrass and with rice. sprangletop are likely to germinate while the field Land leveling and the proper construction of is drained. Appropriate chemical methods are levees permit uniform depth of water and reduce therefore necessary to control weeds resulting weed infestations {10), Level land, which requires from or not controlled by water management prac- fewer levees, reduces the number of sites for weed tices. growth. Annual weed grasses and broadleaf weeds Phosphate or nitrogen fertilizer applications to are usually more abundant on ridges and along ricefields stimulate growth of many weed grasses field margins where water does not cover the land and aquatic weeds {10), Phosphate incorporated adequately. Some emersed aquatic weeds, such as in the soil before dry seeding rice stimulates ducksalad, purple ammannia, redstem, and water- growth of young grass plants. When phosphate is hyssop, grow best in shallow water, and are more applied to a crop preceding rice in the rotation, or abundant in low areas where surface drainage is just before the first irrigation, it contributes less inadequate. Submersed aquatic weeds grow very to weed growth. Early to midseason applications densely in California ricefields flooded to depths of nitrogen are usually advantageous to rice (7). greater than 6 inches, especially in the deep bor- However, if barnyardgrass is present, nitrogen ap- row ditches adjacent to high or wide levees, but plied early stimulates growth of grass and en- these submersed aquatic weeds are not that trou- hances weed competition {16), Herbicides control blesome in leveled fields with shallower water. these weeds and permit timely applicaitons of Herbicides reduce problems with weeds in rice- phosphate and nitrogen. fields that are not level or that have improperly Chemical Control Methods constructed levees. The seeding method influences weed problems. As in other crops, herbicides have become im- Barnyardgrass that emerges along with drill- or portant tools in the cultivation of rice in the United dry-seeded rice is difficult to control by cultural or States to reduce weed losses and to increase the mechanical methods. Water seeding, initiated in effectiveness of other cultural practices {10), California, reduces growth of barnyardgrass and The herbicides 3',4'-dichloropropionanilide (pro- sprangletop during rice emergence. HeAicides are panil) or Ä-ethyl hexahydro-lZT-azepine-l-carbo- essential for control of annual grasses that typi- thioate (molinate), applied early in the season, cally infest dry-seeded rice, or of aquatic, sedge, control young weed grasses and some broadleaf, and broadleaf weeds that infest water-seeded rice. aquatic, and sedge weeds. Broadleaf, aquatic and Water management, before and after seeding sedge weeds that are not controlled by propanil or rice, often determines the kinds of weeds in the molinate, or those that germinate after these her- field and affects the severity of infestations. Flood- bicides are applied, are controlled by follow-up ing the field for water seeding decreases problems treatments of phenoxy herbicides such as (2,4- with weed grasses, but increases problems with dichlorophenoxy) acetic acid (2,4-D) ; [(4-chloro- aquatic weeds. Flooding dry-seeded rice after ö-tolyl)oxy]acetic acid (MCPA) ; (2,4,5-trichloro- emergence to a depth of 4 to 8 inches kills young phenoxy) acetic acid (2,4,5-T) ; and 2-(2,4,5-tri- weed grasses that are 1 to 3 inches tall (16). How- chlorophenoxy) propionic acid (silvex). Each year ever, this early flooding has a narrow margin of about 80 percent of the rice crop is treated with propanil or molinate and about 50 percent with safety to the rice and may also increase aquatic phenoxy herbicides. weeds. Draining the field soon after water seeding PROPANIL.—Barnyardgrass and many other an- helps to control aquatic weeds and algae {10) '^ nual weeds 2 to 3 inches tall are controlled by pro- draining also helps rice seedlings to emerge panil at 3 to 5 pounds per acre ^ with little or no through early growths of American pondweed and to establish better stands during cool weather. *An application rates refer to active ingredients. 138 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE,

injury to rice (P, 11), However, propanil fails to Sequential propanil treatments also control control weeds that emerge after treatment. Pro- weed grasses and aquatic weeds on good soils {H) panil is effective on rapidly growing weeds, but it The first treatment of propanil at 3 to 5 pounds frequently fails to control weeds growing slowly per acre, applied postemergence to drained fields, because of dry soil or low temperatures. Irrigation controls annual grasses 2 to 3 inches tall. Fre- before treatment or delay of the treatment until quently, when fields are flooded within a week after warmer temperatures prevail stimulate the growth treatment, aquatic weeds become a problem. A sec- of weeds and increase their susceptibility. If delays ond treatment of propanil at 3 to 4 pounds per acre, in treatment permit grass weeds to reach the five- applied to drained fields, usually controls aquatic leaf to tillering stages, applications at higher rates weeds less than 2 inches tall. of 5 to 7 pounds per acre are more effective than at MoLiNATE.—In fields to be water seeded, granu- lower rates. lar or emulsifiable molinate applied at 3 pounds Although rice has wide tolerance to treatments per acre and harrowed into the soil before planting with propanil, significant injury to the crop can and flooding provides preemergence control of develop. Applications of propanil within 15 days barnyardgrass. However, in fields to be dry seeded, before or after applications of certain phosphate molinate applied before planting may injure rice insecticides may injure rice significantly. Applica- significantly. tions of propanil after seed treatments or after After emergence of either dry- or water-seeded postemergence treatments of certain carbamate in- rice, 2 to 3 pounds of molinate, per acre applied secticides may injure or kill rice. Likewise, alter- into the water in granular form by aircraft, will ing the water depth rapidly before or after pro- control barnyardgrass that is up to 3 inches tall panil treatments increases the chance of crop (-5, 13), This control of emerged barnyardgrass, injury. however, requires that the irrigation water be held Propanil is diluted with water, and the resultant for at least 7 days after treatment. If the water is emulsion is usually applied at about 10 gallons drained after 7 days, the residual control by per acre by fixed- or rotary-wing aircraft, or at molinate lasts for approximately 2 additional higher volumes of 15 to 40 gallons per acre by weeks ; weed grasses that germinate later are not tractor or hand sprayers {16). Propanil does not likely to be controlled. control weeds residually, but flooding the ricefield Molinate also controls annual spikerushes and to a depth of 3 to 4 inches within 4 days after eclipta, but it is ineffective against most other treatment usually prevents more weeds from ger- broadleaf, aquatic, and sedge weeds. Molinate is minating (P). Control may not be effective if rain more effective than propanil when cool tempera- falls within 8 hours after propanil treatment {11), tures prevail at time of treatment or where the irri- Seqential propanil treatments control bam- gation water is drained soon after treatment {11). yardgrass and other weeds effectively on problem Sequential treatments of propanil and molinate (calcareous, saline, or sodic) soils, thereby permit- frequently control barnyardgrass better than sin- ting water to be managed in ways that do not gle or sequential applications of either herbicide injure rice {H), Irrigation soon after an early alone. The first treatment with propanil controls propanil treatment may damage rice grown on emerged grasses 2 to 3 inches tall. If the irrigation problem soils (5). When the water is withheld to water is delayed more than 7 days after the pro- prevent rice injury, barnyardgrass frequently re- panil treatment, more barnyardgrass may germi- infests the field. The first application of propanil nate and infest the ricefield. These later infesta- at 2 to 3 pounds per acre controls barnyardgrass up tions are controlled with 2 to 3 pounds of molinate to 2 inches tall, and a second treatment at 3 to 4 per acre applied into the water when the barn- pounds per acre controls barnyardgrass up to 3 yardgrass is 2 to 3 inches tall. inches tall that germinated after the first treat- PHENOXY HERBICIDES.—Postemergence applica- ment. Usually the rice is large enough soon after tions of 2,4-D MCPA, 2,4,5-T, or silvex at 0.5 to the second treatment to escape injury from the 1.5 pounds per acre control many broadleaf, aqua- subsequent flooding. tic, and sedge weeds, but do not control weed RICE IN THE UNITED STATES 139 grasses {16), They control weeds effectively when and sedge weeds more effectively than propanil or diluted with water and applied at total volumes of molinate alone. Propanil at 3 to 5 pounds per acre 15 to 40 gallons of spray material per acre by hand applied to drained fields or molinate at 2 to 3 or tractor sprayers or at 3 to 5 gallons per acre by pounds per acre applied to flooded fields controls aircraft sprayers. young weed grass. When aquatic weeds infest rice The stage of growth of rice influences its re- after watering, 2,4,5-T or silvex applied to drained sponse to phenoxy herbicides {16), Phenoxy her- fields at 0.5 to 0.75 pound per acre 2 to 3 weeks bicides at rates necessary for weed control fre- after propanil or molinate treatments controls quently injure young rice plants (seedlings 0 to 3 weeds less than 2 inches tall and usually injures the weeks old) and those in the early-tillering, late- rice only slightly. jointing, booting, or heading stages. Eice in late- COPPER SULFATE.—Copper sulfate (pentahy- tillering to early-jointing stages is usually tolerant. drate) at 0.5 parts per million by weight of cop- Phenoxy herbicides applied at midseason when per (approximately 2 pounds of copper sulfate rice is in the early-jointing stage (basal internodes per 4 acre-inches of water), applied into the irri- 0.5 inch long or less) usually do not injure rice, gation water as soon as algal colonies form on the but they do control weeds effectively. Rice varie- soil, controls green and blue-green algae and does ties in all maturity groups tolerate all phenoxy not injure rice {If), Copper sulfate is more effec- herbicides at this stage of growth. tive on green than on blue-green algae and fails Rice responds differently to specific phenoxy to control weed grasses and broadleaf, other herbicides {16). MCPA, 2,4,5-T, and silvex ap- aquatic, and sedge weeds. It is most effective when plied during the early-tillering to mid-tillering applied to acid soils low in organic matter or to stages of growth injure rice less than 2,4-D. Silvex water low in bicarbonate salts. Bicarbonates in the and 2,4,5-T injure young rice less than MCPA. soil or water inactivate copper as copper carbonate. Any of these herbicides may reduce yields severely Copper sulfate at rates as low as 0.33 parts per mil- if applied during the booting and heading stages. lion by weight may injure fish. If water with cop- Benefits from applications of phenoxy herbicides per sulfate is drained from the ricefield into irri- made in the less tolerant stages of growth of the gation systems, fish, shellfish, and other aquatic rice usually do not exceed the losses of crop injury animal life may be injured. from the treatment except when weed infestations are heavy and the weeds are obviously suppressing Herbicide Drift the development of the rice. Weeds also respond differently to specific Field crops, such as cotton and soybeans, and phenoxy herbicides {16), Hemp sesbania is very certain horticultural, fruit, and ornamental plants susceptible to 2,4-D and 2,4,5-T; northern joint- may be injured by spray drift of propanil or phe- vetch is tolerant to 2,4-D but susceptible to 2,4,5- noxy herbicides (i, ^, 8^ 11), When sufficient her- T ; ducksalad is more susceptible to 2,4-D than to stands and yields may be reduced, and maturity 2,4,5-T. Where several species of weeds that vary bicide gets onto these crops, plants may be injured, in susceptibility to phenoxy herbicides are present may be delayed. The amount of herbicide that in the same field, mixtures of herbicides are often crops will tolerate without damage depends on the more effective than either of these herbicides used crop, growth stage at time of contact, environ- alone. ment, soil, plant nutrition, and other factors. Drop- Sequential treatments of propanil or molinate let size, density of droplets, herbicide formulation, and phenoxy herbicide (2,4,5-T or silvex) control wind speed and direction, height of the spray weed grasses and broadleaf, aquatic, and sedge nozzle above the treated surface, temperature, rela- weeds {14), Propanil or molinate followed by a tive humidity, and other factors all influence drift treatment of either of these phenoxy herbicides of herbicides. Drift of herbicides from the target controls barnyardgrass as effectively as propanil area may be reduced in several ways as described or molinate alone and controls broadleaf, aquatic, previously (^, 16), 140 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

Selected References (8) SMITH, R. J., JR. 1965. EFFECTS OF CHLOROPHENOXY HERBICIDES ON (1) KLINGMAN, G. C. SOYBEANS. Weeds 13: 168-169. 1961. WEED CONTROL : AS A SCIENCE. 421 pp. John (9) Wiley & Sons, Inc., New York. 1965. PROPANIL AND MIXTURES WITH PROPANIL FOR (2) MuziK, T. J. WEED CONTROL IN RICE. Weeds 13: 236- 238. 1970. WEED BIOLOGY AND CONTROL. 273 pp. MC- (10) Graw-Hill Book Co., New York. 1967. WEED CONTROL IN RICE IN THE UNITED STATES. (3) OELKE, E. A., and MORSE, M. D. Asia-Pacific Weed Control Interchange 1968. PROPANIL AND MOLINATE FOR CONTROL OF Proc. 1: 67-73. BARNYARDGRASS IN WATER-SEEDED RICE. (11) Weed Sei. 16: 235-239. 1968. CONTROL OF GRASS AND OTHER WEEDS IN RICE (4) OLSEN, K. L. WITH SEVERAL HERBICIDES. Ark. Agr. 1957. "SCUM" CONTROL IN RICE FIELDS AND IRRIGA- Exp. Sta. Rpt. Ser. 167, 38 pp. (12) TION CANALS. Ark. Agr. Exp. Sta. Rpt. 1968. WEED COMPETITION IN RICE. Ser. 69, 20 pp. Weed Sei 16: 252-255. (5) PLACE, G. A. (13) 1969. RELATIONSHIP OF IRON, MANGANESE, AND 1970. MOLINATE FOR BARNYARDGRASS CONTROL IN BICARBONATE TO CHLOROSIS OF RICE GROWN RICE. Weed Sei. 18: 467-469. ON cALCERous SOIL. Ark. Agr. Exp. Sta. (14) Rpt. Ser. 175, 21 pp. 1970. SYSTEMS FOR WEED CONTROL IN RICE. Rice (6) SEAMAN, D. E., MORSE, M. D., MILLER, M. D., and Jour. 73(4) : 11-18. others. (15) and FRANS, R. E. 1968. CONTROLLING SUBMERSED WEEDS IN RICE. 1969. HERBICIDE MANAGEMENT IN RICE AND SOY- Calif. Agr. 22 (11) : 11. BEANS ROTATIONS. (Abstract) Weed Sei. Soc. Amer., 1969 meeting, Abs. No. 21. (7) SIMS, J. L., HALL, V. H., and JOHNSTON, T. H. (16) and SHAW, W. 0. 1967. TIMING OF N FERTILIZATION. I. EFFECT OF AP- 1966. WEEDS AND THEIR CONTROL IN RICE PRODUC- PLICATION NEAR MIDSEASON ON VARIETAL TION. U.S. Dept. Agr., Agr. Handb. 292, PERFORMANCE. Agron. Jour. 59: 63-66. 64 pp. RICE DISEASES

By J. G. ATKINS

Eice diseases are economically important in been one of the chief factors in preventing suc- the rice crop in the southern rice area (3). Ex- cessful rice culture in southern Florida. cept for seedling disorders and stem rot, these dis- When the fungus attacks the leaves, the dis- eases are of no importance in California. Annual ease results in leaf blast. Symptoms are elongated losses in Louisiana and Texas average about 8 per- or spindle-shaped leaf spots with grayish centers cent. Diseases are important in rice, but they are and brown margins. Under severe disease condi- not as disastrous in rice as in many other crops. tions, nearly all leaves are killed and many young Although average losses in most fields are rather plants are killed or severely damaged. Young low, losses in individual fields from specific dis- plants arc also killed by infection of the sheath eases are sometimes high. tissues, which turn brown. Generally, all plants in Most of the world's important rice diseases are a riccfield are not uniformly affected. In some areas known to occur in the United States. Exceptions the plants are killed, but in others they are affected are some virus diseases, bacterial leaf blight, and less severely. The leaves are susceptible from the certain physiological disturbances reported from seedling through the early-tillering stage of Japan and other Asiatic countries. A few minor growth. Under flooded soil conditions, the leaves diseases caused by fungi are also not known to are less susceptible in the late-tillering stage of occur in the United States. growth and at heading. As a result, severely dis- The severity of specific diseases is influenced by eased fields of young rice appear to recover later varietal susceptibility, fertilization, soil type, and when the plants become older and progress through environmental conditions. Losses from certain the vegetative growth stages. diseases can be reduced or held to a low level by Head blast, also called rotten neck, results from the use of resistant varieties, better cultural and the attack of the fimgus after emcr-gencc of the management practices, and seed treatment. Dis- panicle (head) from the boot or top leaf sheath. eases shift in prevalence and severity with The peduncle (the top node and internode) and changes in varieties. The reactions of 18 rice the branches of the panicle are very susceptible to varieties to some diseases in the United States are shown in table 19.

Major Diseases Blast Blast, caused by the fungus Pyricularia ory- zae Cav., occurs in all ricegrowing areas of the world and is the most damaging disease of rice (fig. 39). Blast in the United States is about as old as the rice crop itself—it caused trouble shortly after rice was introduced into South Carolina [25) and then into Louisiana {19). AN Losses from blast were light from about 1935 to BN-22014 FIGUR?: 39.—Long narrow spots on leaves and infection at 1955; however, in the southern rice area, par- at base of panicle, which caused it to break over, are ticularly Louisiana, seriously infested fields have typical symptoms of blast. (Photo courtesy of Arkansas been found each year since 1955 {9). Blast has Agricultural Experiment Station.) 141

88-871 O - 73 - 10 142 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

TABLE 19.—Reaction oj rice varieties to some diseases oj rice in the United States [R = resistant; MR = moderately resistant; MS = moderately susceptible; S = susceptible; VS = very susceptible]

Grain type and variety Blast Brown leaf Straighthead White tip Narrow brown Hoja blanca spot leaf spot

Short-grain : Caloro -- S ._ S___ _ S . S - S . S. Colusa -- S ._ MS- _ MS . S . S . R.

Medium-grain : Arkrose -- S ._ S___ - S - VS . S . MR. Calrose -- vs ._ MS_ _ S . vs - MS S. CS-M3 ._ vs ._ MR. _ s . vs . R Nato -- s ._ S-__ - MS . s - MS S. Nova 66 - R .- s.._ - MR . s . MR R. Saturn .. R ._ S-__ _ S R . MS S. Zenith .- R, S ._ MS- - S._-._ S . S S.

Long-grain : Belle Patna .- S ._ S-__. . R R . S s. Bluebelle .- s ,_ S-__. - R R . s s. Bluebonnet 50 - MS _ S-._. . R R . s s. Dawn - R _ MR_ . VS R MS s. Delia .- S _ MS_. . s R . MS s. Rexoro .- S _ S___. . vs R . VS s. Starbonnet .- s _ s___. . MR R S s. Toro .- s _ s___. . MS R MS s. TP 49 ,_ s _ s__.. . VS R MS s.

attack. Typically, the peduncle shows a necrotic, knolls or other high areas in the ricefield than it is brown area that prev^ents movement of food into on ÛIQ low areas. the developing grain. The grain produced in Nearly all United States rice varieties are sus- affected panicles varies from nearly normal ceptible to one or more of the pathogenic races amounts to none, depending on the time of infec- of P, oryzae known to occur in the southern rice tion in relation to flowering. As infection weakens area (^, i^, 22), However, Dawn, Nova 66, Saturn, the structural tissues of the peduncle, the panicles and Zenith are resistant to the most prevalent frequently break over to give the condition known races. Rexoro and TP 49 usually escape infection as rotten neck. because they are late maturing. Under average field Atmospheric moisture conditions are of pri- conditions, Bluebonnet 50 is less severely infected mary importance in infection and spread of the than several other varieties, and losses are lighter. fungus that causes blast (^0), Frequent rains, Other rice varieties listed in table 19 are consist- heavy nightly dews, and high relative humidities ently susceptible. The reaction of 18 varieties to favor disease development. Severe outbreaks of five International races of P. oryzae {12^ 23) is the disease occur after periods of rainy weather. given in table 20. Rice plants under high levels of nitrogen fer- tilization are susceptible to blast. Plants grow- Brown Leaf Spot ing under nonflooded or upland conditions are Brown leaf spot of rice is caused by Helmin- more susceptible than those growing under flooded thosporium oryzae B. de Haan, the same fungus or irrigated conditions (20). For this reason, blast that causes seedling blight. The spots occur chiefly is often heavier on the levees and on or around on the leaves but are frequently on the hulls (fig. RICE IN THE UNITED STATES 143 40). The leaf spots are oval to circular and grayish cease to grow normally and turn yellow. Affected brown to black. plants generally show heavy brown leaf spot be- Brown leaf spot is prevalent in the southern cause of their impaired physiological condition. rice area. Under good cultural conditions, dam- Several soil fungi may cause root rot. Damage age is slight on vigorous plants. Weak plants with is more severe when nematodes and root mag- yellowish leaves, caused by low nitrogen levels, gots feed on the roots. Plants growing in saline root damage, or other factors unfavorable for good soil and alkali spots in fields are also affected. nitrogen nutrition, are often severely damaged by Losses from root disorders can be held to a brown leaf spot. minimum through good cultural and fertilizing All commercial rice varieties are susceptible to practices that maintain the plants in a vigorous brown leaf spot. Bluebonnet 50 is generally more condition. Draining the field and permitting the susceptible than are most other varieties. Manage- soil to dry stimulates growth of new roots. For ment and fertilizer practices that promote good many years growers have used this practice to con- growth reduce losses. trol root maggots and straighthead in conjunction with topdressing applications of nitrogen. Root Rot Brown-Bordered Leaf and Sheath Spot Eoot rot of rice is a general diseased condi- tion in which the roots grow poorly, darken with Brown-bordered leaf and sheath spot in rice is necrosis, or die. As decay progresses, the leaves caused by Rhizoctonia oryzae Ryker & Gooch. During most seasons, the disease appears when the plants approach maturity. It usually is most severe in grassy areas where the foliage is dense {28). The disease has increased in importance as va- rieties and cultural practices have changed and rates of nitrogen fertilizer are higher. The di- sease reduces yield and milling quality throughout the Southern States. The disease affects plants in small circular areas (fig. 41). These spots may coalesce so that there are large areas where the plants are killed. The reddish-brown spots on the leaves and sheaths are

iJ^^n^-niÇ^ô^^y

BN-21974 PN-2786 FIGURE 40.—Oval to circular lesions on rice leaves are FIGURE 41.—Dead rice plants caused by brown-bordered symptoms of brown leaf spot. leaf and sbeatb spot at harvesttime. 144 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

TABLE 20.—Reaction of rice varieties to International races of Pyricularia oryzae Cav. in the United States [R = resistant; MR=moderately resistant; S = susceptible]

Grain type and variety Race IB-5 Race IB-49 Race IB-54 Race ID-13 RaceIG-1

Short-grain: Caloro S R_ s. Colusa R_ s. Medium-grain : Arkrose R_ S-. s. Calrose R_ S_. s. CS-M3 S_ R_ s_. s. Nato S_ R_ s_. s. Nova 66 S_ R_ s_. R. Saturn S. R_ s_. R. Zenith S. S_. R_ R. Long-grain: Belle Patna S__-. S-._. s_. s.. s. Bluebelle S___. S___. s_. s_. s. Bluebonnet 50 S S__-. s_. s_. s. Dawn MR_ MR_ R_ R_ R. Delia S S___. s_. s_. s. Rexoro S_ s_ s. Starbonnet S S S_ s_ s. Toro S S_ s_ s. TP49 S_ s. s.

large, 2 to 6 centimeters long, and may eventually Stem Rot cover the entire plant. Generally the medium- and short-grain varieties are less susceptible than the The stem rot fungus MagTiaparthe salvinii long-grain varieties. (Catt.) Krause & Webster (21) lives in the soil as sclerotia from one season to another up to 6 years Seedling Blight (34) (fig. 43). Typically, rice plants are attacked in the advanced-growth stages. Initially, the fun- Fungi in the soil and on or in the seed cause gus invades the sheath tissues near the water level seedling blight of rice (fig. 42). These fungi and produces dark areas. The fungus progresses reduce emergence and kill or weaken the plant inward to the culm. The nodes and intemodes turn after emergence. Low soil temperature and high dark and weaken. Many of the plants lodge be- soil moisture, combined with seedborne fungi, or cause of rotten culms. The plants are weakened combinations of the two, make stand establish- or killed before maturity. Little grain is produced ment difficult. and its quality is reduced. Satisfactory stands are generally obtained None of the rice varieties are highly resistant through use of good-quality seed treated with a to stem rot (15). Eexoro is perhaps the most sus- fungicide and seeded under conditions favorable ceptible variety. Some of the medium-grain varie- for rapid emergence. Atkins, Cralley, and Chilton ties are resistant or intermediate in reaction. The (7) reported that a number of chemicals effec- early-maturing varieties tend to escape severe tively control seedling blight. Seedling blight damage if seeded early. High rates of nitrogen caused by Helminthosporiwn oryzae—the fungus fertilizer increase the susceptibility of plants to that causes brown leaf spot—can be partly con- stem rot damage, but potassium fertilizers reduce trolled by fungicides. damage. RICE IN THE UNITED STATES 145 the soil with water. Undecayed organic material and arsenic in the soil are also factors. In general, rice grown on the lighter or sandy soils is more subject to straighthead than rice grown on the heavier clay soils. For many years, ricegrowers have followed the practice of draining and drying the soil to pre- vent straighthead (31). The fields are drained, dried, and reflooded just before panicle formation. This period is about 50 days after seedling emer- gence for Century Patna 231, a susceptible, early- maturing variety {14-)- However, the best control measure is to use resistant varieties. The numerous rice varieties have been classified as to straighthead reaction (6). Belle Patna, Blue- bonnet 50, Nova 66, Saturn, and Starbonnet are resistant. Toro and Nato are intermediate in reac- tion and are seldom affected.

White Tip

White tip of rice is caused by an ectoparasitic, foliar nematode, Aphelenchoides hesseyi Christie (fig. 45). The nematodes are seedborne and live from one crop to the next in the seed rice. After rice is sown, the nematodes become active and move into the growing point of the young rice plants. In this protected location, the nematodes feed and re- produce. The feeding by large numbers of nema- todes injures the developing leaves and panicles before emergence. The injury is later observed as

FIGURE 42.- -Three blighted rice seedlings and, at right, a white, necrotic leaf tips and small, usually sterile healthy seedling.

^BLA Straighthead

Straighthead, sometimes called blight, is a non- parasitic or physiological disease of rice. Diag- nostic symptoms are observed only in the panicles. The panicles remain upright at maturity because of lack of grain development (fig. 44) ; hence, the common name of straighthead. The shape of the palea and lemma, which later form the hulls, is distorted, particularly in the long-grain varieties. It is similar to a parrot beak, crescent, or half moon. Hull distortion may or may not be conspicu- ous in the short- and medium-grain varieties. PN-2997 FiGURE 43.—Rice culms showing sclerotia of stem rot Straighthead is caused by unknown soil condi- fungus. (Photos courtesy of ArkansasII Agricultural tions associated with prolonged submergence of Experiment Station.) 146 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

eral grains per panicle are affected. Other plant parts are not affected. The disease is easily over- looked in the field. It can be seen when rains wash the black spores over parts of the pani- cle or when the smut spore mass expands between the two hulls with absorption of moisture from dew or rain. The black smut spores germinate and produce sporidia. When the rice ñowers open at pollina- tion, the sporidia enter. The fungus then invades the developing grain, grows, and produces its spores. Infection is not systemic in the rice plant, as it is with several other cereal smut fungi {27). Although the smut spores are seedbome (that is, they are carried into a field along with the seed), they are not directly responsible for later smut infection. Kernel smut is more preva- lent in rainy seasons and in fields receiving fairly high rates of nitrogen fertilizer {29). Nitrogen BN-22036 applied late in the season also increases the inci- FIGURE 44.—Erect panicles with sterile florets are typical dence of kernel smut. of rice plants affected by straighthead. No satisfactory controls for kernel smut are known at present. Seed treatment by chemicals panicles. Grain yields in susceptible plants are or hot water is ineffective. All long-grain varie- greatly reduced. ties are susceptible. Most of the medium-grain Several methods can be used to control white tip. Resistant varieties or nematode-free seed can be sown. Cralley {16) and Todd and Atkins {3, 32) have described hot-water treatments that control white tip very effectively. They are not practical for general use. However, they can be used by the rice experiment stations for treating small lots of base seed used in foundation seed programs. Emer- gence through water following water seeding con- trols white tip {17). This practice can also be used to clean up small lots for seed production. Several rice varieties are resistant to white tip (table 19). In general, the long-grain varieties are resistant and most of the short- and medium-grain varieties are susceptible {13). Although white tip was an economically important disease in the south- ern rice area for many years, it is no longer important.

Kernel Smut

Kernel smut of rice is caused by one of the smut fungi, Neovossia 'barclayana Bref. {35).

Part or all of the endosperm is replaced by a FIGURE 45.—Left, healthy rice panicle; right, panicle black mass of smut spores (fig. 46). One to sev- affected by white tip. RICE IN THE UNITED STATES 147 more white stripes along the leaf blade or a whiten- ing of the entire leaf blade (fig. 48, J.). Diseased plants are generally reduced in height. The pan- icles are reduced in size and often fail to emerge completely from the boot. The palea and lemma are generally distorted in shape and later turn brown. Most ñorets are sterile. Because the dis- eased plants produce few, if any, seed, the panicles remain upright instead of bending over like those of normal plants (fig. 48,5 ). Hoja blanca occurs only in the Western Hemi- sphere. Leaf symptoms are similar to those of the stripe virus disease of Japan, but the two diseases differ {8). Hoja blanca was first recognized as a new rice disease in 1956 {ß). Severe yield losses were reported from Cuba, Venezuela, and other countries of . The disease and insect vector were found in 1957 in Florida {5), in 1958 EN-22020 in southern Mississippi {10), and in 1959 in Louisi- FIGURE 46.—Rice panicle with several florets aiïected by kernel smut. ana {11). An eradication program was initiated after each finding of the insect vector. In 1960 and 1961, neither the disease nor the insect vector was varieties, except Nato, generally are less suscep- found in the southern rice States. In 1962, S. orizi- tible than the long-grain varieties. cola was again found and collected in southern Louisiana {1). Minor Diseases

Narrow Brown Leaf Spot Narrow brown leaf spot in rice, caused by the fungus Gercospora oryzae Miyake, is one of the most prevalent rice diseases in the southern rice area. The leaf spots are narrow or linear and light to dark brown (fig. 47). When the disease is severe, the leaves die, one after another, begin- ning with the lower leaves. In most seasons, infec- tion does not become severe until late August or September. Early-maturing rice varieties, when sown early, tend to escape heavy infection. Kice varieties show marked differences in sus- ceptibility. Several pathogenic races or strains of C. oryzae occur in the United States {26). Bluebonnet 50 and Rexoro are susceptible varie- ties. Under average field conditions. Nato, Toro, and TP 49 are fairly resistant.

Hoja Blanca

Hoja blanca, or white leaf, is a virus disease of BN-21981 rice that is transmitted by a planthopper, Soga- FIGURE 47.—Linear lesions on rice leaves are symptoms of todes orizicola Muir. Leaf symptoms are one or narrow brown leaf spot. 148 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE 30, 33), rice was binder harvested and placed in shocks. The change in harvesting methods from the binder to the combine has probably reduced losses from kernel spots.

Leaf Smut

Leaf smut, caused by Entyloma oryzae H. «S: P. Syd., is a common but minor rice disease in the southern rice area (fig. 49). The disease may be .„ recognized by numerous, small, black sori on the leaves. The disease becomes more prevalent as the rice plants approach maturity.

BN-21973

'I I. n I I

f ' . • !•

't ' t if! n , ' 4 ■ '•If

BN-2197( Í FIGURE 48.—Hoja blanca on (A) rice leaves and (B) rice ( ♦ 1" • panicles. 1, I '. if J' I*,Il ■ ! I. Yield losses from hoja blanca in the United ■J States have been negligible. Unless the disease and I insect vector become established and cause eco- I'.'«. ♦ nomic losses, no control measures need be consid- ered by individual rice producers. Arkrose, Colusa, and Gulfrose are resistant to hoja blanca. • i':. .,j li l< Kernel Spots FIGURE 49.—Leaf smut on rice leaves. Discolored or dark areas and spots occur on the kernels of brown and milled rice. This kernel dis- coloration is generally associated with dark, dis- Selected References colored hulls. Curvularia lunata (Wakk.) Boed., FusariuTTi spp., Altemaria spp., Tnchoconis cau- (1) ANONYMOUS data (Appel & Strunk) Clemments, and Helmin- 1962. INSECT CARRIER OF HOJA BLANCA RICE DIS- thosporium oryzae are the fungi most commonly EASE REAPPEARS IN LOUISIANA. Rlce JOUr. 65(12) : 22. isolated {33). Feeding punctures of the immature (2) ADAIR, C. R., and INGRAM, J. W. grains by stink bugs also cause kernel spots (18). 1957. PLANS FOR THE STUDY OF HOJA BLANCA—A At the time of some of the earlier studies (^4) NEW RICE DISEASE. Rice Jour. 60(4) : 12. RICE IN THE UNITED STATlEiS 149

(3) ATKINS, J. G. (17) 1958. RICE DISEASES. U.S. Dept. Agr. Farmers' 1956. A NEW CONTROL MEASURE FOR WHITE TIP. Bul. 2120, 14 pp. Ark. Farm Res. 5(4) : 5. (18) DOUGLAS, W. A., and TULLíS, E. C. (4) 1962. PREVALENCE AND DISTRIBUTION OF PATHO- 1950. INSECTS AND FUNGI AS CAUSES OF PECKY GENIC RACES OF PIRICULARIA ORYZAE IN THE RICE. U.S. Dept. Agr. Tech. Bui. 1015, 20 UNITED STATES. Phytopathology 52: 2. pp. (19) FULTON, H. R. (5) and ADAIR, C. R. 1957. RECENT DISCOVERY OF HOJA BLANCA, A NEW 1908. DISEASES AFFECTING RICE IN LOUISIANA. LA. RICE DISEASE IN FLORIDA, AND VARIETAL RE- Agr. Expt. sta. Bui. 105, 28 pp. SISTANCE TESTS IN CUBA AND VENEZUELA. (20) KAHN, R. P., and LIBBY, J. L. U.S. Dept. Agr. Plant Dis. Rptr. 41: 911-915. 1958. THE EFFECT OF ENVIRONMENTAL FACTORS AND PLANT AGE ON THE INFECTION OF RICE BY (6) BEACHELL, H. M., and CRANE, L. E. 1956. REACTION OF RICE VARIETIES TO STAIGHT- THE BLAST FUNGUS, PIRICULARIA ORYZAE. HEAD. Tex. Agr. Expt. Sta. Prog. Rpt. Phytopathology 48: 25-30. 1865, 2 pp. (21) KRAUSE, R. A., and WEBSTER, R. K. 1972. THE MORPHOLOGY, TAXONOMY, AND SEXUAL- (7) CRALLEY, E. M., and CHILTON, S. J. P. 1957. UNIFORM RICE SEED TREATMENT TESTS IN ITY OF THE RICE STEM ROT FUNGUS, MAGNA- ARKANSAS, LOUISIANA, AND TEXAS. U.S. PORTHE SALVINII (LEPTOSPHAERIA SALVINII). Dept. Agr. Plant Dis. Rptr. 41: 105-108. Mycol. 63: 103-114. (22) LATTERELL, FRANCES M., TULLíS, E. C, and COLLIER, (8) GOTO, K., and YASUO, S. 1961. COMPARATIVE REACTIONS OF RICE VARIETIES J. W. TO THE STRIPE AND HOJA BLANCA VIRUS 1960. PHYSIOLOGIC RACES OF PIRICULARIA ORYZAE DISEASES. Internl. Rice Comn. Newsletter CAV. U.S. Dept. Agr. Plant Dis. Rptr. 44: 10(4) : 5-8. 679-683. (9) JoDON, N. E., and BOLLICH, C. N. (23) LING, K. C, and Ou, S. H. 1962. TESTING AND BREEDING FOR BLAST-RESISTANT 1969. STANDARDIZATION OF THE INTERNATIONAL RICE. La. Agr. 5(4) : 14-15. RACE NUMBERS OF PYRICULARIA ORYZAE. Phytopathology 59: 339-342. (10) KRAMER, J. P., and HENSLEY, S. D. 1958. HOJA BLANCA AND ITS INSECT VECTOR FOUND (24) MARTIN, A. L., and ALSTATT, G. E. ON RICE IN A SECOND AREA IN THE UNITED 1940. BLACK KERNEL AND WHITE TIP OF RICE. TeX. STATES. U.S. Dept. Agr. Plant Dis. Rptr. 42 : Agr. Expt. Sta. Bui. 584, 14 pp. 1414. (25) METCALF, H. 1906. A PRELIMINARY" REPORT ON THE BLAST OF RICE (11) NEWSOM, L. D., SPINK, W. T., and others. 1960. OCCURRENCE OF HOJA BLANCA AND ITS VEC- WITH NOTES ON OTHER RICE DISEASES. S.C. TOR, SOGATA ORIZICOLA MUIR, ON RICE IN (26) RYKER, T. C. LOUISIANA. U.S. Dept. Agr. Plant Dis. Rptr. 1943. PHYSIOLOGIC SPECIALIZATION IN CERCOSPORA 44: 390-393. ORYZAE. Phytopathology 33: 70-74. (12) -ROBERT, A. L., ADAIR, C. R., and others. (27) TEMPLETON, G. E. 1967. AN INTERNATIONAL SET OF RICE VARIETIES 1961. LOCAL INFECTION OF RICE FLORETS BY THE FOR DIFFERENTIATING RACES OF PIRICULARIA KERNEL SMUT ORGANISM, TILLETIA HóRRIDA. ORYZAE. Phytopathology 57: 297-301. Phytopathology 51: 130-131. (13) and TODD, E. H. (28) and JOHNSTON, T. H. 1959. WHITE TIP DISEASE OF RICE. III. YIELD 1969. BROWN-BORDERED LEAF AND SHEATH SPOT ON TESTS AND VARIETAL RESISTANCE. Phyto- RICE. Ark. Farm Res. 18(1) : 5. pathology 49: 189-191. (29) JOHNSTON, T. H., and HENRY, S. E. (14) CHEANEY, R. L. 1960. KERNEL SMUT OF RICE. Ark. Farm Res. 1955. EFFECT OF TIME OF DRAINING OF RICE ON THE 9(6) : 10. PREVENTION OF STRAIGHTHEAD. Tex. Agr. (30) TISDALE, W. H. Expt. Sta. Prog. Rpt. 1774, 5 pp. 1922. SEEDLING BLIGHT AND STACK-BURN OF RICE (15) CRALLEY, E. M. AND THE HOT-WATER SEED TREATMENT. U.S. 1936. RESISTANCE OF RICE VARIETIES TO STEM ROT. Dept. Agr. Dept. Bui. 1116, 11 pp. Ark. Agr. Expt. Sta. Bui. 329, 31 pp. (31) and JENKINS, J. M. (16) 1949. WHITE TIP OF RICE. (Abstract) Phyto- 1921. STRAIGHTHEAD OF RICE AND ITS CONTROL. pathology 39: 5. U.S. Dept. Agr. Farmers' Bui. 1212, 16 pp. 150 AGRICULTURE HANDBOOK NO. 289, U.S. DEPT. OF AGRICULTURE

(32) TODD, E. H., and ATKINS, J. G. (34) and CEALLEY, B. M. 1959. WHITE TIP DISEASE OF RICE. II. SEED TREAT- 1941. LONGEVITY OF SCLEROTIA OP THE STEM-ROT MENT STUDIES. Phytopathology 49: 184- FUNGUS LEPTOSPHAERIA SALVINII. Phyto- ,oo^ n. l^^'r. pathology 31: 279-281. (33) TULLíS, E. C. (35) . ^„^ JOHNSON, A. G. 1936. FUNGI ISOLATED FROM DISCOLORED RICE KER- 1952. SYNONYMY OP TILLERTIA HÓRRIDA AND NEO- NELS. U.S. Dept. Agr. Tech. Bui. 540, 11 VOSSIA BARCLAYANA. Mycologia 44- 77*- PP- 788. INSECTS AND THEIR CONTROL

By JAMES R. GIFFORD

Rice Wafer Weevil because it is unreliable and also impractical in areas where water supplies are not abundant. The rice water weevil {Lissorhoptrus ory- Carbofuran effectively controls the rice water zophilus Kuschel) occurs in most of the rice- weevil. growing areas of the United States. The adult Rice Stink Bug weevil (fig. 50) is grayish brown and about one- eighth inch long. It overwinters in bunch or finely Practically all ricefields in Arkansas, Louisi- matted grasses growing in the vicinity of ricefields. ana, and Texas are infested with the rice stink In the spring it migrates into the ricefields and bug {Oebalus pugnax (Fabricius)). The adult feeds on newly emerged plants, making slitlike, stink bug (fig. 51) is a straw-colored, shield- longitudinal feeding scars on the upper surface shaped insect about one-half inch long. It passes of the leaves. Eggs are laid in the leaf sheaths the winter as an adult in the refuse near the sur- of rice plants. Larvae feed on the roots of rice face of the ground in clumps of grass. It emerges plants and cause severe injury by pruning the from winter quarters in the spring and feeds on root system. The young larvae, or root maggots grasses and sedges. Two or three generations may as they are commonly called, are milky white, leg- be produced there before it migrates to rice, soon less, and about one-half inch long when fully after the rice begins to head. The eggs, shaped like short cylinders, are deposited on the leaves, grown. Since the immature stages of the rice water stems, or heads of the rice, on grass, on mexican- weevil are spent entirely under water among the weed, and on other weeds. The eggs, usually laid rice roots, many of the larvae can be destroyed in clusters of 10 to 40 in two rows, are green when by drainage {6). However, this practice is not first laid, and change to reddish black before hatch- recommended for controlling the root maggots ing. The freshly hatched nymph is nearly round, about 11/2 millimeters long and 1 millimeter wide. The head, thorax, legs, and antennae are black; the abdomen is red with two elongate black spots running crosswise. The young stink bugs remain together near the eggshells until after the first molt. Then they seek separate feeding places, gen- erally on grass or on rice panicles, where they suck the juice from the developing kernels. Feed- ing in the milk stage of rice produces empty glumes, whereas feeding in the soft-dough stage causes "peckiness" of grain or seed sterility (^). Since infestations build up on grasses earlier in the season, control of grasses in the field and mowing the areas around the field help reduce the potential population for infesting rice. Two hymenopterous ^gg parasites, Ooencyrtus anasae (Ashm.) and Telenomus podisi Ashm., are at times very helpful in reducing stink bug popu-

PN-2999 lations late in the season. Malathion, methyl para- FIGURE 50.—Adult rice water weevil. thion, and carbaryl control the rice stink bug. 151 152 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

young rice plants and destroy the leaf tissue. Infested leaves turn brown and lie prostrate on the water. This pest is seldom present in suffi- cient numbers to warrant control measures. Grasshoppers are found in nearly all ricefields, but seldom are present in sufficient number to cause severe damage {2) They feed on the leaves, culms, and grain of rice. Sporadic outbreaks of the fall armyworm {Spodoptera frugiperda (J. E. Smith)) (fig. 52) have been recorded at irregular intervals and in widely separated localities. It feeds on leaves and stems of unflooded rice. Submerging the rice crop is usually effective in controlling the fall army- worm. The rice stalk borer {Ghilo plejadellus Zincken) (fig. 53) and the sugarcane borer {Diatraea sac- charalis (Fabricius) ) (fig. 54) both occur in rice- fields in Louisiana and Texas {S). These borers tunnel inside the stem throughout the growing season and interfere with normal growth and de- velopment of the plant. They may also weaken the stem so that it breaks off or lodges before

FIGURE 51.—Adult rice stink bug.

Other Pests of Rice

Several other pests occasionally damage rice severely. The grape colaspis {Maecolaspis -ßavida (Say)), called lespedeza worm by ricegrowers, sometimes reduces plant stands severely in Arkan- sas and Louisiana {9). It passes the winter in the young larval stage. In the spring the larvae make their way toward the soil surface where they feed on germinating seeds or seedlings. Heavy infestations may reduce stands so severely that re- seeding is necessary. The larvae pupate in the soil and emerge as pale-brown, elliptical beetles about one-eighth inch long. Adults of the grape colaspis lay their eggs in the soil around the roots of grasses growing in lespedeza or other leguminous crops. Eice planted after these crops is subject to damage by the colaspis. The rice leaf miner {Hydrellia griseola var. FIGURE 52.—FaU armyworm : {A), Male moth ; (5), right scapulasis Loew) is a pest of rice in California front wing of female moth; {C), moth in resting posi- tion ; {D), pupa ; and {E), full-grown larva. A, B, D, E, (7). Maggots of this fly feed in the leaves of about X 2 ; (7, slightly enlarged. RICE IN THE UNITED STATES 153 that has become the scourge of rice production in several Latin American countries. 8. orizicola was first reported in the United States in 1957. Al- though the planthopper was found in Mississippi in 1958 and in Louisiana in 1959 (i) and in 1962, neither it nor the disease it transmits has become established in the United States. Apparently the vector cannot survive the cold weather that some- times occurs in the rice-growing areas of this country. The chinch bug {Blissus leucopterus (Say)) (fig. 55) is present in Arkansas, Louisiana, and Texas {3). It has entered ricefields in large num- bers and seriously injured the young rice plants before they were submerged. Both adults and nymphs attack rice. The feeding of bugs causes

P.\-3003 the plants to wither and die. Chinch bugs feed FiGUBE 53.—Rice stalk borer: (A), Adult and (ß), larva. mainly on the stems, just above the surface of the About X3. ground. They may be controlled by submerging the infested field. The insects spend the winter in dry grass, straw, and other material that affords them shelter. Plowing under such material in the fall or winter reduces the number of bugs emerg- ing the following spring. The tadpole shrimp {Triops longicaudatus (LeConte)), although not an insect, is sometimes a pest in California (JO). Damage occurs shortly after fields are flooded and when eggs laid in fields the previous year hatch. Shrimp larvae at first feed on organic matter in the soil ; but as they mature, FIGURE 54.—Adult female sugarcane borer. they dislodge and feed on young rice plants. The shrimp matures in 8 to 10 days and may produce a harvest. Stalk borer damage is first noticeable second generation. on the rice plant at time of heading when sterile heads with white panicles, or white heads as they are commonly called, appear. The eggs of both species of borers are parasitized by the minute wasp (Trichogramma minutum Riley), which helps reduce insect populations. Plowing under rice stubble in the spring destroys some of the overwintering borers. Grazing with cattle or ñood- ing rice stubble fields reduces the number of hiber- nating borers. Eicefields should be separated as far as possible from corn and sugarcane because these two crops serve as a breeding place for the sugar- cane borer. The planthopper {Sogatodes orizicola Muir) is potentially a very important pest of rice. It is the only known vector of hoja blanca, the rice disease FIGURE 55.—Chinch bugs. Enlarged. 154 AGRICULTURE HANDBOOK NO. 2 89, U.S. DEPT. OF AGRICULTURE

For information on the insecticides currently (4) and TULLíS, E. C. recommended for control of rice insects, consult 1950. INSECTS AND FUNGI AS CAUSES OF PECK Y RICE. U.S. Dept. Agr. Tech. Bui. 1015, 20 your county agent, State agricultural experiment pp. station, or the U.S. Department of Agriculture, (5) GRIGARICK, A. A.. LANGE, W. H., and FINFROCK, Washington, D.C. 20250. D.C. 1961. CONTROL OF THE TADPOLE SHRIMP, TRIOPS

LONGICAUDATUS, IN CALIFORNIA RICE FIELDS. Jour. Econ. Ent. 54: 36-40. (6) IsELY, DWIGHT, and SCHWARDT, H. H. Selected References 1934. THE BICE WATER WEEVIL. Ark. Agr. Expt. Sta. Bui. 299, 44 pp. t (1) ATKINS, J. G., NEWSOM, L. D., SPINK, W. T., and (7) LANGE, W. H., JR., INGEBRETSEN, K. H., and DAVIS, others. L.L. 1960. OCCURRENCE OF HOJA BLANCA AND ITS INSECT 1953. RICE LEAF MINER. Calif. Agr. 7(8)': 8-9. VECTOR, SOGATA ORIZICOLA MUIR, ON RICE IN (8) PORTMAN, R. F., and WILLIAMS, A. H. y LomsiANA. U.S. Dept. Agr. Plant Dis. Rptr. 1952. CONTROL OF MOSQUITO LARVAE AND OTHER 44: 390-393. PESTS IN RICE FIELDS BY DDT. JoUr. EcOU. (2) BOWLING, C.C. Ent. 45: 712-716. 1960. CONTROL OF RICE STINKBUGS AND GRASSHOP- (9) RoLSTON, L. H., and ROUSE, PHIL. 1960. CONTROL OF GRAPE COLAPSIS AND RICE WATER PERS ON RICE. Tex. Agr. Expt. Sta. Prog. WEEVIL BY SEED OR SOIL TREATMENT. Ark. Rpt. 2132, 6 pp. Agr. Expt. Sta. Bui. 624,10 pp. (3) DOUGLAS, W. A., and INGRAM, J. W. (10) ROSENBERG, L. E. 1942. RICE FIELD INSECTS. U.S. Dept. Agr. Cir. 1947. APUS AS A PEST IN CALIIORNIA RICE FIELDS. 632,32 pp. Calif. Dept. Agr. Bui. 36(2) : 42-48.

U.S. GOVERNMENT PRINTING OFFICE : 1973—0-488-871