Cotton Boll Weevil: An Evaluation of USDA Programs : a Report (1981)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

l. COTTON CULTURE AND COTTON INSECT PESTS COTTON AND COTTON CULTURE IN THE UNITED STATES The genus Gossypium, to which the cotton plant belongs, may be divided into 2 groups, according to chromosome number. One group has a haploid chromosome number of l3, the other 26. Commercial cotton belongs mainly to the group with 26 chromosomes that originated in Central America. This group consists mainly of the cultivated spe- cies G_^ hirsutum and G._ barbadense. Cotton is generally an annual plant when cultivated, although it can also be grown as a perennial in warm temperate zones. Cotton is cultivated in a wide range of soil types that are well-drained and aerated but retentive of moisture. It can be cultivated between 43° north latitude and 25° south latitude. Successful cultivation requires a mean warm season temperature of over 24°C, an annual rainfall of 400 to l,200 mm with favorable seasonal distribution, abundant sunshine during the period of boll maturation and harves- ting, and a frost-free period longer than l60 days. The maturation of the boll from the time of fertilization to the time of splitting takes about 50 days. The mature lint is formed by extrusions of the epidermal cells of the testa (seed coat). These are long tubular cells with a heavy cellulose wall. The staple length of the mature lint ranges from l to 5 cm or more in different species and varieties. Cotton has long been a major crop in the United States. Along with tobacco and rice, cotton was an agricultural export of colonial America, and its importance increased during the nineteenth century to the point where the cultivation of cotton became a dominant influ- ence in the social and economic development of the southern states. Ebeling (l980) has described the social and economic changes that accompanied the expansion of cotton culture in the South. Cotton culture survived the social and economic upheavals of the Civil War, although cotton growers in the southeastern states often suffered economic depression and depletion of soil nutrients even- tually resulted in widespread malnutrition in the region (Ebeling l980). At the beginning of the twentieth century cotton was still

the major agricultural crop in the Old South, in recent years/ however, cotton cultivation in the southwestern states and California has become important due to the implementation of irrigation projects (Ebeling l980). The boll weevil, which appeared in the United States in the last decade of the nineteenth century, was the first important insect pest of cotton (Parencia l978). The search for measures to control the boll weevil and the development of other significant pest problems have become overriding considerations in cotton cultivation. Recent control technology has included the widespread use of a varied arsenal of synthetic organic chemicals. The history of insecticide development, use, and abuse, and the concomitant changes in cotton cultivation are discussed in later sections of this chapter. COTTON PRODUCTION Cotton production in the United States has ranged from l7 to 22 percent of world cotton production since l974. Russia, China, and the United States together produce about one-half to two-thirds of world cotton. While most cotton seeds are used domestically, about half of U.S. cotton lint is exported. These exports vary sharply from year to year, in contrast to the relatively steady domestic lint consumption of 6 million bales per year. In l979-l980, for example, 9.2 million bales of U.S. cotton were exported, accounting for more than 40 percent of world cotton trade; in l980-l98l less than 6 million bales were exported, about 30 percent of world cotton trade. Cotton comprised 6.3 percent of the total value of U.S. agricultural exports in l979 and 6.8 percent in l980. Cotton accounts for half of the fiber used worldwide and for one-quarter to one-third of U.S. fiber consumption. It is an impor- tant crop to the world, the United States, the textile industry, the economies of the cotton states, and to several hundred thousand people employed in cotton-related industries. Cotton's share of the world fiber market has been trending downward, however, and will probably continue to fall, reaching 40 percent by the year l990 (Figure l.l). Nonetheless, per person fiber use has been rising about 4 percent a year. Future world fiber consumption may grow 2.5 to 3 percent a year and world cotton use l.3 to 2 percent a year (Figure l.2). Worldwide cotton use rose from 45 million bales in the early l960s to 65 million in l980-l98l and by l990 may be 80 million bales (Collins et al. l979). The upward trend in use of cotton (Figure l.3) is relatively smooth compared to the upward trend in worldwide production (Figure l.4). Production of cotton in the developing and Communist (Central Plan) countries has increased faster than in the United States. Between l964 and l977 the largest increases and the greatest rates of increase in both use and production were in the Communist countries (Collins et al. l979), and the greatest future expansion in cotton use is expected to be in developed countries outside the United States (Collins et al. l979). Data for l977 through l980 show that

10 PEI 80 *CENT — 60 ^<'^^^ — 40 • Equivalent Alternative II A 20 — — g I ! i I II II II I I I I I II II I I I I 1960 '63 '66 '69 72 75 78 '81 '84 '87 '90 A THE LOWER SHARE PROJECTIONS ARE THE SHARES IMPLIED BY DIVIDING THE TOTAL FIBER USE PROJECTIONS OF ALTERNATIVE I INTO THE COTTON USf PROJECTIONS OF ALTERNATIVE II. FIGURE l.l Actual and projected levels of cotton's share of the total fiber market. Cotton's share of the world fiber market declined from 68% in l960 to 50% in l974. Alternative I & II (dotted line): projects cotton to have 43% in l990. Equivalent of Alternative II (dashed line) projects lower share by dividing the more optimistic total fiber use by the least optimistic projection of cotton production. SOURCE: Collins et.val. (1979) MIL. METRIC TONS 40 30 20 16 Alternative I ^ —• " World •^~~~' Alternative 1964'66 '68 70 72 74 '85 '90 FIGURE l.2 Actual and projected annual average growth rates of world fiber use. World fiber use grew about 4% per year from l964-l974. Alternative I projects 3.l% growth to the year l990 by a more rapid growth of world population and fiber use per capita than Alternative II. SOURCE: Collins et al. (1979)

l1 MIL BALES 80 70 60 50 40 24 20 16 World 2.2% Alternative I + 2.0% _^ — —' _ ^00» ^^ __ ^^ •*• * "**^«» ^** ^^ ^^ + 1 3% Alternative II Developing -0.2% I I I I Alternative I + 1 .4% __ _ — — ir^r 09% Alternative 28 24 20 16 __ Central Plan Alternative I 1.3% Alternative II 24 r— 20 16 12 1964 '66 '68 70 '72 '74 Alternative + 2.8% ^ ^ ~~ + 1.9% Alternative '85 '90 FIGURE l.3 Actual and projected annual average growth rates of cotton use. World cotton use expanded 2.2% per year up to l974 which was only half the rate of expansion of total fiber use. Alternative I projects 2% growth to l990 by anticipating slower growth of total fiber use and a declining share for cotton. Alternative II anticipates the same market share for cotton but lower cotton yields, fewer acres, less people, and lower use of fiber per capita. SOURCE: Collins et al. (1979)

12 Alternative I _ _ - -I- 2.1 % ^ ^ — — "J^—— .^^-—*"-" Alternative II +1.3% i I I I I It I I I 30 26 22 30 26 22 18 14 16 14 12 10 8 1964 '66 '68 70 72 74 76 Alternative I ,.. — — ' + 2.4% Alternative II + 1.3% Central Plan Alternatives I, II ^ —• • — — *~ *" + 1.5% Alternative I +1.8% -0.1% Alternative II '85 '90 FIGURE l.4 Actual and projected annual average growth rates of cotton production. Expansion of world cotton production may speed up in the future (Alternative I) in spite of slower growth in Central Plan Countries. Under Alternative II slower yield growth and expansion of acreage is expected in all regions. SOURCE: Collins et al. (l979)

l3 cotton use in the Communist countries is expanding more rapidly than expected, while production is expanding less rapidly than expected (note dots, Figures l.3 and l.4). Use of cotton in the United States has declined since the early l960s. Total use of all fibers in the United States grew from 36 pounds per capita in l960 to 56 pounds in l980, while domestic cotton use fell from 23 to l4 pounds per capita. Cotton's share of the fiber market fell from 64 to 27 percent between l960 and l980 but may decline more slowly to 22 percent by l990 (Collins et al. l979). It appears that the extent of the change to man-made fibers may be nearly complete. Table l.l gives the cotton production for the major cotton producing states in the USA. There has been a steady shift of cotton acreage within the United States toward the West. This shift is primarily the result of better cotton-growing conditions—a longer growing season in Arizona and California, a greater number of large flat fields in Texas, and less damage from pests in both areas. These advantages have resulted in lower per bale production costs in the West than in the Southeast or in the hills of the Delta states. In addition, western cotton has usually been sold at slightly higher prices because of its better quality. In l977 yields were above normal for the Delta and the Southwest and below normal for the Southeast, and the cost differences were very large. Production costs per pound of lint in l977, excluding land costs, were 50 cents in the West, 46 cents in the Southwest, 56 cents in the Delta, and 99 cents in the Southeast. Regional prices for cotton in l977 were relatively low (56 cents in the West, 50 cents in the Southwest, 52 cents in the Delta, and 52 cents in the Southeast), but these region- al price differences reflected the typical pattern—cotton in the West averaged 4 cents above the national average of 52 cents, and cotton in Texas averaged 2 cents below. Up to l974, various government programs retarded the westward movement of cotton. Prior to l97l, marketing quotas discouraged expansion in the West. The quotas were eliminated in l97l, but a program involving a minimum price of l5 cents per pound, with bonuses for small farmers, was maintained until l974. When the small farmer bonuses were eliminated, cotton acreage dropped in the Southeast (see Figure l.5) . If the trends shown in Figure l.5 continue until l990, the reduction in acreage may be slower in the Southeast and the Delta, and expansion in Texas and Oklahoma may be slower. Some additional expansion of irrigated cotton acreage may take place in Arizona and California, although the increase in production costs associated with irrigation may limit this expansion and spur the trend to dryland cotton in Texas and Oklahoma. The economic return from using water to irrigate cotton—for example, in the Imperial Valley of Califor- nia—is generally higher than from water used to irrigate grain crops, but it is lower than the return from irrigating off-season vegetable, vine, and tree crops. National average yield per acre may not rise as much in the future or may even fall if there is a reduction in the Southwest's

l4 TABLE l.l Average cotton acreage, yield and production l975-l979 for l0 major cotton producing states and the total for the United States. Upland Cotton Acres harvested Yield (lbs. Production (l000 acres) per acre) (l000 bales) Texas 5593 347 4ll3 California l295 957 25l5 Mississippi l232 526 l330 Arizona 483 l0l9 l020 Arkansas 770 475 753 Oklahoma 463 33l 330 Louisiana 477 55l 545 Alabama 36l 4l9 3ll Tennessee 289 378 222 Missouri 209 447 l88 Total U.S. ll,643 48la ll,75l Average yield per acre. SOURCE: USDA (l98lb)

15 36 25 15 Southwest A . I i I Soutb«Mt YT-W i I 1960 '62 '64 '6 A fBCLIMIHAHY 70 72 74 76 IB" *80 FIGURE 1.5 U.S. harvested cotton acreage by regions. The shift of cotton toward the West accelerated after 1974. The share of cotton in the West outside the boll weevil belt is 17% of acreage but is 35-40% of production. SOURCE: Collins et al. (1979) ability to irrigate cotton. An expansion of Texas and Oklahoma dryland cotton production may be large enough to offset any produc- tion declines in the Southeast and West. In the Delta states, which produce 20 to 30 percent of the cotton produced in the United States, future acreage and yield levels are not projected to change significantly. No significant change is expected in these states over the long run in the price ratio between cotton and soybeans, and only moderate increases are expected in rice production. Cotton acreage and production have fallen in the Southeast. During the early l960s the amount of acreage used for cotton in this area was about 2.5 million, or one-sixth of U.S. total. In the early l970s it declined to l.5 million acres, and was about 600 thousand acres, or one-twentieth of the U.S. total from l978 to l980. Cotton production in the Southeast fell from about 2 million bales, or 13

l6 percent of the U.S. total, in the early l960s to about l.3 million bales, or l0 percent, in the early l970s. Between l978 and l980 it dropped to 0.6 million bales, or 5 percent. Cotton production in the Southeast has been dramatically restructured as well. In l960 some 80 to 90 percent of the cotton there was picked by hand. By l970 more than 95 percent was picked by machine. Cotton production in the Southeast has a relatively high gross value of $200 to $400 per acre, which is higher than soybeans ($l50 per acre) or corn ($200 per acre). Cotton production costs in the Southeast, however, are a high $300 per acre. The average net land rent per acre of cotton is often not larger than the average net land rent for corn and soybeans. Soybeans are less costly per acre and provide a more certain profit. Cotton will probably continue to be competitive with corn and soybeans in the Southeast and the Delta only on larger fields with better soil or in times when the supply of cotton is relatively short. The trend in the application of insecticides to cotton fields is downward. Cotton growers now have a strong economic incentive to assess the size of insect populations and their potential damage before applying pesticides. The use of scouting procedures allows beneficial arthropods to control cotton insects as long as possible before the application of insecticides. Increases in the costs of insecticides and of applying them, relative to the costs of scouting and pest management consultant services, are leading to the adoption of different techniques. THE ECOSYSTEM CONCEPT A growing awareness of the cottonfield as part of the ecosystem has begun to influence cotton pest insect control. Although control programs are still based primarily on insecticide use, systems of pest management utilizing other techniques have begun to be developed. The concept of integrated crop management or agroecosystem management is emerging. The holistic concept of ecosystems, which has become a dominant theme in environmental biology (Odum l97l), has also assumed importance in modern agronomy (Todd l98l). The major precepts of holism are the interconnection of different parts of an ecosystem, beyond the sum of the system's parts, and the existence of homeo- static (self-regulatory) mechanisms. An agroecosystem has been defined as "a unit composed of the total complex of organisms in a crop-producing area together with the overall conditioning environment and as further modified by various agricultural, industrial, recreational and social activities of man" (Smith and Reynolds l972). Nutrient cycling processes are altered by the harvesting of agroecosystems. They are also altered by consumer organisms—insects, weeds, and other pests—within the ecosystem, which are controlled with pesticides; by tillage, which incorporates organic matter into the soil, thus increasing decomposition rates and hastening nutrient release from decomposing organic matter; and by the addition of fertilizer nutrients, which often cannot be stored

17 efficiently by the system and are partly lost via leaching. Although agricultural management causes large-scale changes in agroecosystems, ecosystem-level processes continue to operate. The Cotton Agroecosystem Extensive reviews of the cotton agroecosystem and its components have been published by Smith and Reynolds (l972) and Reynolds et al. (l975). These reviews discuss in detail the plant system, the complex of invertebrate plant pests, soils, water, fertilizer, and weather that affects the plant system, and the various human influ- ences. The cotton plant has undergone considerable artificial selec- tion. The varieties currently grown have been selected for fiber quality and yield, insect resistance, fruiting characteristics, and other traits which permit, or may require, changes in cultivation and management. Like many other agroecosystems, cotton fields may con- tain a variety of weed species. Weeds are considered undesirable because they compete with crop plants for sunlight, water, and fer- tilizer and may serve as alternate hosts for insect pests. But weeds may also serve as refuges for predaceous and parasitic arthropods of cotton plant pests (Altieri and Whitcomb l979), and nutrient uptake by weeds may serve to retard nutrient loss from the cotton ecosystem. The insect complex of cotton fields is a varied one, including not only pest species and their predators but many other arthropod species. The presence of weeds in cotton fields and in surrounding vegetation doubtless has a major influence on the numbers of arthro- pod species in the cotton fields and may sustain a desirable diver- sity of arthropod species. Attempts to manage consumer organisms, however, are aimed more frequently at suppressing them than at promoting their predators or parasites. Future management practice may include using invertebrates as regulators in the decomposition process, although current practice does not attach importance to the decomposition process as a way of recycling nutrients. Nutrients, and frequently water, are provided by artificial methods. Although no-tillage practices conserve soil fertility, using them for cotton probably is not practical at present because of insect pests and disease problems. The most significant difference between the abiotic environments of cotton and natural ecosystems is the number and volume of synthetic organic chemicals— insecticides, herbicides, defoliants—added to cotton ecosystems. The variety of cotton plants, the timing of planting, and the density and size of fields are other human-controlled variables. The cotton ecosystem may be the most chemically altered of all our agroeco- systems. COTTON INSECTS Cotton fields contain a surprisingly varied and complex popula- tion of insect pests and entomophagous (insect-eating) organisms,

l8 such as birds, small mammals, and other insects. There is general agreement among entomologists about the importance of entomophagous species for maintaining optimum cotton production over the long term. Key Cotton Insect Pests According to Reynolds et al. (l975), there are at least one or two key pests of cotton wherever the crop is grown in the United States. There is also a large array of insect and spider mite species that are considered to be "occasional" or "potential" pests. Damaging populations of the occasional pests occur sporadically, some become problems only when their natural enemies are eliminated by insecticides. Adkisson (l973) listed the key pests in the three major cotton-producing regions in the United States as follows: • West: pink bollworm (Pectinophora gossypiella), tobacco budworm (Heliothis virescens), and plant bugs (Lygus species). • Southwest: boll weevil (Anthonomus grandis), fleahopper (Pseudatomoscelis seriatus), bollworm (Heliothis zea) , and tobacco budworm. • Delta and Southeast: boll weevil, plant bug (Lygus lineo- laris), bollworm, and tobacco budworm. There is a small area in Arizona subject to boll weevil infesta- tion which is non-contiguous to the main boll weevil region. In the last decade overwintered "stub" cotton which, in effect, lengthens the season, has provided great impetus for this displaced infestation to occur (Bergman et al. l98l). The boll weevil is considered the key insect pest in the Missis- sippi Delta despite the current situation, in which the bollworm and the tobacco budworm create larger losses. Prior use of insecticides to control boll weevils accounts for the current problems (Parvin et al. l977, Clower l980). The tobacco budworm did not become a serious pest until the synthetic organic pesticides were introduced in the l940s. Between l972 and l978 H_^ virescens was the dominant species of Heliothis from August to October in the Delta. The widespread alteration of the cotton field environment by insecticides has also turned other minor pests into major problems. The cotton leaf perforator (Bucculatrix thurberillea) population has reached serious proportions in recent years. Whitefly (Trialeurodes abutilonea) was not a problem until damaging populations developed in the Red River Valley in Louisiana in the late l960s. Whiteflies now extend eastward as far as Georgia (Jones et al. l976). Beneficial Arthropods In the absence of insecticides, populations of arthropod preda- tors of cotton insect pests normally become large and diverse.

19 Whitcomb and Bell (l964) found over 600 predaceous species of in- sects, mites, and spiders in Arkansas cotton fields. Gonzalez et al. (l977) reported that total predator populations are generally more abundant in California than elsewhere, which may be a result of the state's numerous alfalfa fields. When the alfalfa is cut about every 30 days, vast numbers of predators are driven out and migrate to surrounding fields. Van den Bosch and Hagen (l966) have published a detailed list of the parasitic and predaceous species more commonly found on cotton. The overall value of parasites in controlling cotton insect pests is not known, in part because of the complexity of faunal relationships. Attempts have been made to establish biological control of the boll weevil, pink bollworm, and lygus bugs by import- ing exotic parasites, but these have had no measurable effect, in fact, some of the imported parasite species have failed even to become established. No pathogens that are effective against the boll weevil have been reported.