Division of Agricultural Scien c e $ UNIVERSITY OF CALIFORNIA ^€) "•;;'«.",:"< ■ ■ ■ ■■■:: : ' . •-.-' ■ v ;-,-,--- ■■■ ■--■ - — : .-— -'••"- WEED CONTROL -;f : „/v. 1LLER • FOY • KEMPEN • CARTER • HOOVER CALIFORNIA AGRICULTURAL EXPERIMENT STATION BULLETIN 791 1 Annual grass in cotton at end of irrigation run. DECEMBER, 1962 THE PROBLEM Cotton, California's leading cash crop, returns growers a total of approximately $300,000,000 annually from the state's*- 800,000 planted acres. Despite this, cot- ton has a serious production problem — that of weed control. Weeds reduce the^ value of the cotton crop from 4 to 6 per cent each year, and growers in the San ' Joaquin Valley report that they spend about $20 an acre yearly in attempting ■ to control them. I Weed damage often is the result of several factors. Early weeds may com-^l pete with cotton for moisture, sunlight, and nutrients, thus lowering production and increasing hand-labor costs. Heavy I THE AUTHORS: John H. Miller is Agronomist, Crops Research Division, Agricultural Research Service, U. S. Department of Agriculture; Chester L. Foy is Lecturer in Botany and Assistant Botanist in the Experiment Station, Davis; Harold M. Kempen is Assistant Agricul- turist, Agricultural Extension Service; Lyle M. Carter is Agricultural Engineer, Agri- cultural Engineering Research Division, Agricultural Research Service, U. S. Depart- ment of Agriculture; Marvin Hoover is Agriculturist, Agricultural Extension Service, U. S. Cotton Field Station, Shafter. NOTE: This is a cooperative contribution from the University of California Agricultural Experi- ment Station and Extension Service and the Crops Research and Agricultural Engineering Research Divisions, ARS, USDA. Submitted for publication May 4, 1962. THE COVER PHOTO was taken at the University of California's West Side Field Station, near * Five Points. It shows test plots used in the study of the effectiveness of herbicides on cotton. Similar work is being carried on in other cotton-growing areas of the state by University and USDA scientists. infestations of weeds late in the season may interfere with cotton defoliation, reduce the efficiency of both mechanical r and hand pickers, and lower the grade of lint by introducing additional stains and trash. ¥ It is estimated that over 90 per cent of California's cotton is now machine- harvested, as compared with about 10 „per cent in 1948. During this same period, hand-labor requirements for cot- ton production were reduced by one- third, and present hand-labor require- ments are largely for weeding. Thus, weed control is the principal barrier to complete mechanization of California's 'cotton crop. While California cotton yields have been high, profits from them have not kept pace with increased production costs. Higher machinery costs and higher wages for hand labor have helped cause a tightening of profit margins. Further, the supply of farm workers can no longer be depended upon to handle Cali- fornia's cotton crop adequately, despite increased wages. If California is to maintain its present high standing in cotton production it is imperative that sound, economical weed control practices be followed. Such prac- tices include the best combination of cultural, biological, and chemical methods. [3] CULTURAL PRACTICES Long term weed-control programs are advisable. Early elimination of small in- festations, even at high cost, results in economy when considered over a long- term period. Consistent yearly efforts are essential in eliminating remnants of a weed population. Dangers of re-infesta- tion from dormant seeds (such as field bindweed, or wild mustard), or from ad- jacent weedy areas, irrigation water, or grazing animals, must be considered be- fore the eradication program is initiated. Frequently, a community eradication program is justified, although full partic- ipation by all parties is required for suc- cess. After eradication of the selected weed species other species can be included in the control program until the entire problem is minimized or eliminated. After analysis of the problem, a long term program can be started using any of the following methods of eliminating weeds: flooding, competitive cropping, hoeing, cultivation, burning, chemical control, fallow, systematic rotation — or various combinations of these. Crop rotation *• Good crop rotation is an important part of any well planned weed-control program. For cotton, the rotation should x include crops whose growth habits, cul- tivation, and management interfere greatly with growth of troublesome ^ weeds. Annual weeds are restricted, and often eliminated, by a rotation system which includes a small grain crop, such ^ as barley, a cultivated crop, such as cot- ton, and a pasture or meadow crop. Cer- tain other weeds are held in check by rotations which include smother crops, such as alfalfa. The weed problem may also be less- ened by practices which prevent weeds ^ from reseeding, by a cultivation schedule that interferes with the normal develop- ment of most weeds, or by a selective - chemical weed-control program for crops preceding cotton. When crops having similar growth habits are grown for successive years on the same field an environment favorable to at least one ' type of weed is established. Definition of Terms Used Layby: A term used to describe the time of the last cultivation. This time varies with growers, but is generally about mid-July in the San Joaquin Valley and mid-June in the desert valleys. Carrot oils: Examples are Richfield Selective 1, Shell 10, Stoddard Solvent, Analos 7, Coberly & Plumb Herbicidal Oil 1. Dirting: The act of moving soil into the plant row during cultivation, either to cover weeds or to build the bed profile for furrow irrigation. By-pass agitation: A method of maintaining an emulsion or suspension of insoluble compounds by returning and recirculating a portion of the sprayer-pump output to the spray tank. Irrigation run: Synonymous with the length of irrigation furrow. psi: pounds per square inch (gauge). LP: liquid petroleum. Rhizomes: Underground stems of perennial weeds capable of regenerating new plants vegetatively. Also known as "rootstocks." x* [4] Proper selection of rotations often per- mits the maximum use of chemical aids in weed control. For example, 2,4-D may be used to control annual weeds in barley preceding cotton, whereas the reverse order of planting would be im- possible. Also, diuron may be used both in cotton prior to cotton, corn or sor- ghum, and in alfalfa before cotton — but not in most other crop combinations, be- cause of injurious residues left in the soil. Finally, flame cultivation is a satis- factory approach in cotton, but is out of the question in barley. Land preparation and planting The starting point for a weed-control program in cotton is land preparation. Proper land leveling eliminates low spots where water stands after each irrigation; this is important because water trapped in low areas of the field, or at the ends of irrigation "runs," provides excellent locations for the establishment of weeds. Irrigation systems which provide for ef- ficient utilization of water accumulated at the end of the field will reduce weed problems. Thorough incorporation of old plant residues into the soil is also important for efficient early weed control. Where large amounts of plant residue are pres- ent, shredding followed by efficient cov- ering by plowing affords better seed-bed conditions and allows early effective cul- tivation. Good planting practices result in more uniform emergence of cotton — a pre- requisite to successful early cultivation. Good soil moisture and soil tilth are im- portant considerations for uniform emer- gence. The uniformity of cotton emer- gence has also been markedly improved in recent years through use of the seed- firming wheel on planters, and better seedling disease control with fungicides. Where heavy rates of seeding are used, mechanical thinning of cotton will reduce the hoe-time required for early weeding and thinning. Equipment adjustment Use of accurately adjusted implements is highly advisable in such operations as bedding, planting, and cultivation, al- though the adjustment of equipment in the fields is often difficult and time-con- suming because of irregularities in the soil surface. A line diagram similar to the one shown will aid materially in proper adjustment and requires only a smooth, level, surface marked with paral- lel lines representing the plant rows and the middles between the rows. Although other smooth surfaces will suffice, a con- crete slab makes a permanent surface which can easily be marked with painted lines. For field rows 40 inches apart, parallel lines should be spaced 20 inches apart. The equipment can then be placed on the surface with the wheels centered exactly over lines representing the mid- dles. The implement attachments to be used are then set as desired in relation to lines representing the plant rows. The line diagram should be large enough to handle multiple-row equipment. When implements are properly pre-set, only minor adjustments are necessary in the field. Early cultivation Mechanical cultivation is still the prin- cipal method of weed control used in western irrigated cotton. A variety of implement attachments can be used as supplements to normal sweep cultivation for early weed control. Research has shown that reversed disk hillers, bed knives, and rotary hoes, utilized indi- vidually or in combination, are effective tools for weed control early in the season when only limited amounts of soil may be moved. Only small seedling weeds may be killed easily; therefore, proper timing of the tillage operation is essen- tial for effective control. The above-mentioned operations may be begun as soon as sufficient cotton has emerged to enable the operator to follow the row. The rotary hoe attachment for [5] cultivation with the tines reversed has proved to be an effective shield for pro- tecting small cotton. The rotary hoe holds the clods out of the cotton row, yet per- mits "dirting" (covering of small weeds with soil) of very small cotton plants. While most early-cultivating operations are rather slow and tedious, the use of the reversed rotary hoe permits cultiva- tion of very small cotton at normal trac- tor speed. A rubber weeding tool that operates directly under the soil surface offers considerable promise as a device for removing seedling grass weeds in the cotton row. The rubber weeder should not be used on cotton less than about 2 inches tall, but must be used before the secondary root system of the weedy grasses has developed; this tool is best adapted to light soils. Many growers rely exclusively upon mechanical cultivation supplemented by hand hoeing for weed control throughout the season. When cot- ton plants have attained sufficient height to permit it, dirting with sweeps or other mechanical tools is one of the most corn- Below: the line diagram — a useful device for adjusting equipment. Distances between lines equal one-half of row spacing. Furrow Plant row Tractor wheel — furrow Plant row Tractor center line — furrow Plant row Tractor wheel — furrow Plant row - Furrow* [6] mon methods of weed control. If this is done, the bed profile should not be built up so high early in the season that dirt- ing at "layby" is precluded. Combinations of cultivation tools for early weed control. Top left, reversed disk hillers (center) and sweep cultivating tools (left and right); Top right, sweep cultivating tools (top) and bed knives. Lower left, rotary hoe (center) and sweep cultivating tools; Lower right, sweep cultivating tools, rubber weeding tools (center), and bed knives. Table 1. Effect of Date of Last Culti- vation With High-Clearance Equip- ment on Cotton Yields and Picker Efficiency During 1951-1953 at Shafter, California Last cultivation date Yield of cotton Picker efficiency bales per acre 2.71 2.71 2.66 per cent 95.8 95.8 Sept. 1 95.1 * Check plots. Source: Tavernetti, James R., and H. F. Miller Jr., Studies on mechanization of cotton farming in Cali- fornia. University of California Exp. Sta. Bui. 747. If ■-J [7] Late-season cultivation California growers normally layby (cease cultivating) their cotton with a heavy dirting some time in July. Since cotton is usually irrigated after layby, late weed problems frequently develop. Tests with high-clearance equipment (table 1) have indicated that with late- season cultivation a single 20" to 24" sweep in each furrow will provide weed control without injuring the cotton crop. Cultivation should be shallow (not over 2" deep) in order to minimize root pruning, and equipment must be shielded to prevent breaking of branches when traveling at normal tractor speeds. HERBICIDAL OILS Selective herbicidal oils have been suc- cessfully used for controlling seedling weeds in cotton 2 to 6 inches tall. These oils are specially formulated petroleum- naphthas containing no fortifying agent; they generally possess a boiling range of 300° to 400°F and a prescribed per- centage of aromatic components ranging from 16 to 25 per cent. (Consult local Parallel-acting shoes with oiling nozzles attached. agricultural authorities and oil manufac- turers for further specifications and details of proper application.) When these oils are properly used they will kill small weeds without injuring cotton plants. At present, the only selective oils generally available in the western irri- gated cotton areas are known as "carrot oils"; oils of higher boiling range and aromatic content do not exhibit the re- quired selectivity. Selective oiling is specialized: timing, stage of plant growth, rate of application, and spray placement, as well as the prop- erties of the oil, are important in the selective action. For example, oils must be kept off the sensitive cotton terminals, and must not be used when cotton stems begin to crack and form bark. When oils can be used safely, they are applied as laterally-directed sprays at the rate of about 5 gallons per acre of cotton; ap- plication is confined to an 8 to 10-inch band centered on the row. Treatment may be repeated at weekly intervals as needed, but should not exceed three ap- plications. Control of small emerging weeds will generally be excellent where proper care is exercised, although selec- tive oiling will not control perennial weeds, or puncture vine more than 1 to 2 inches in diameter. The biggest problem in oiling is pre- venting injury to the seedling cotton, but this can be avoided by using special application techniques. The first require- ment is smooth, clod-free beds. Oiling operations have been more successful on relatively high-bed plantings, which per- mit establishment of irrigation furrows without moving soil (and additional weed seeds) into the oiled bands. The use of special parallel-acting shoes with shields increases spraying precision and helps to hold soil out of the treated band. Since the two spray nozzles are mounted on the shoes, they follow the contour of the tops of the beds. Best dis- tribution of spray is provided by flat, fan-type nozzles, mounted 1 inch above the bed and directed to the rear and at an angle of about 30 degrees to the row, with a spray pattern directed slightly downward so that the outer edge of the pattern strikes the soil 6 to 8 inches from the nozzle; the oil is applied to a band 8 to 10 inches wide centered on the plant row. This nozzle arrangement per- mits the overlapping of the spray patterns in the cotton row and has given satisfac- tory weed control with a minimum of crop injury. Nozzles should be set so that the spray fan strikes the young cotton stem no higher than 1 inch from ground level. FLAME CULTIVATION Principles and practice Flame cultivation is often an effective supplement to mechanical and chemical weed control methods. Cotton, like most plants, will be killed by the heat intensity used in flame cultivation if it is in the seedling stage. Therefore, practical flame cultivation control of most weeds in cotton depends upon cotton being larger than the weeds. Destruction of weeds in cotton is also influenced by several variables, such as burner design, intensity and volume of the flame, length of exposure, and loca- tion of the burner with respect to the plant. Flame cultivation, like directed oil sprays, eliminates weeds by contact. If viable seeds are present, successive popu- lations of weeds will emerge after each irrigation, and repeated flame cultiva- tions are necessary to control them. Deflection of flame pattern caused by surface irregularities. [9] Although recent burner changes and improved application methods indicate that flaming may sometimes be started when cotton plants are only a few inches tall, initial flame treatments are normally applied with standard-size burners when cotton stems are approximately 3/16 inch in diameter at ground level; the plants are usually about 8 inches tall at this stage of development. With shielded, high-clearance tractors, flame cultivation can be continued until cotton bolls begin to open. In general, flame should be used 3 to 6 days after an irrigation, and this often co- incides with the time when furrows have dried enough to permit use of a tractor; most weeds emerging during an irriga- tion cycle do so during this period. For most effective use of flame, the weeds should not be over 2 inches tall. Flame cultivation, which can be an integral part of a total weed control pro- gram, is used to control only the weeds in the cotton row. As with other weed control practices, uniformity in all field operations will determine the degree of success with flaming. The row profile best suited to flame cultivation is smooth, free of depressions and ridges, and as free as possible of large clods. Laboratory and field obser- vations have shown that row irregulari- ties can cause flame to be deflected up into the cotton plant and cause plant injury; in addition, weeds behind ridges or in depressions will not be touched by the flame. When proper pre-flaming cul- tivations have left the row profile rela- tively smooth, flame cultivation has been found to be equally successful on high rounded, low rounded, or flat beds. Two general systems of pre-flaming practices have been found to be com- patible with flame cultivation in row- irrigated fields: (1) relatively high-bed planting with early weed control pro- vided by application of directed oil sprays, and (2) relatively low-bed plant- ing with moderate to heavy dirting to control weeds prior to the first flaming. When flaming follows directed oil appli- cations, cultivation is normally limited to the furrow, leaving the bed undisturbed. When dirting is used as early-season weed control, it is essential that the final bed profile be developed before the first application of flame. Because soil prob- ably contains weed seeds, none should be thrown to the plants after flaming starts. Data obtained at many locations since 1945 have indicated that flame cultiva- tion, with the adjustment described Table 2. A 10- Year Summary of the Effects of Flame Cultivation on Yields of Cotton Year Date of first flaming Number of flamings Yield Flame plots Check plots 1949 6/7 7/8 6/5 6/13 6/2 6/3 5/3 6/5 6/4 6/3 9 3 5 6 5 5 5 6 3 3 bales per acre 2.21 2.56 2.80 2.19 3.13 2.66 2.22 2.16 3.02 2.46 bales per acre 2 11 1950 2 63 1951 2 65 1952 2 13 1953 3 08 1954 2 66 1955 2 24 1956 2 17 1957 3 06 1958 2 50 Average 6/8 5 2.54 2 52 [10 Table 3. Control of Barnyardgrass in Cotton With Flame Cultivation, and Effects on Grade of Cotton ; 3-Year Summary Grass control* Cotton grade after treatment Treatment grade indexf grade reduction 1.8 6.4 8.4 91.0 97.5 97.8 1 grade Flame cultivation 1/10 grade 1/7 grade * Grass control rating scale: = no control; t Average grade index: 100 = middling; 85 = low middling. herein, will not damage the cotton stands or yields. Table 2 gives a 10-year sum- mary of flame-cultivation test results under relatively weed-free conditions. There were no differences in yields be- tween flamed plots and check plots when fields were flamed three to nine times during a growing season. A 3-year test program designed to de- termine effectiveness of flame cultiva- tion on annual-grass control and reduc- tion of grass in the cotton bale showed it to be more effective than standard sweep cultivation. When an early appli- cation of directed oil spray was used with later flaming, a further increase in weed control was noted (table 3). 10 = perfect control (grass free). 94 = strict low middling; Flaming equipment The components of a typical flame cultivator used in irrigated cotton are: a liquid petroleum (LP) tank, heat ex- changer (fuel vaporizer), controls, fuel lines, and burners. The fuel tank, all fittings, hoses, and tubing used in a flame cultivator must meet certain code specifications of the state and the American Society of Me- chanical Engineers (A.S.M.E.) for han- dling butane and propane fuels. The fuel tank must be stamped with the state in- spector's approval, and the safety valve should be in good operating condition at all times and should be repaired or re- placed only by a competent serviceman. Quick shutoff valve tilth bypass or/fice-i Pressure gauge Pressure regulator r Valve ■*- Heal exchanger— Approved L P Gas Tank Water flow To tractor cooling system Fuel manifold □_!_□ n □ I 1 I 'I I Burners Flame cultivator components. [ii] CH Although the vapor valve is used at the top of the fuel tank to supply gas to burners in some 2-row cultivators, va- porization is not adequate for 4-row equipment. The normal practice is to withdraw liquid from the tank and va- porize it before it reaches the burner nozzles. Two systems of vaporization, known as vapor-burner and liquid- burner are commonly used. In the vapor- burner system, the liquid petroleum fuel is vaporized in a heat exchanger which utilizes the hot water of the tractor's cool- ing system. After vaporization, the gas is conducted through a pressure regu- lator to the manifold and then chan- nelled to the individual burners. In the liquid-burner system, the liquid fuel passes through the pressure regulator and is vaporized in a heat-exchange chamber built into the burner. In field comparisons of flame culti- vators, no differences in weed killing could be found between the two systems of vaporization. During early flame cul- tivations, however, the vapor-burner sys- tems have the advantage inasmuch as they can be quickly turned off when passing over areas where cotton is too small to withstand flame. The time elapsed in turning off the liquid-burner system is longer, as liquid in the lines must be vaporized and burned before the flame is extinguished. Controls should be within easy reach of the tractor operator. A quick-opening valve with a small hole drilled in the gate will allow a pilot light to burn while the burners are shut off as the equipment turns at the ends of the field. Flame burners are normally mounted solidly on the front or rear of cultivating equipment, and burner height is adjusted with the cultivator controls. Front mount- ing permits continuous visibility of the burner pattern and the cotton row dur- ing flaming. Rear mounting places the flame farther away from the tractor fuel tank and permits the tractor operator to drive away from the heat rather than into it. When operating in fields with uniform furrows on land leveled for ir- rigation, solid mounting to the cultivator permits precise vertical control of the burners, and also allows easy turning where ditches and furrows must be crossed. In certain areas of the Cotton Belt where cotton is planted without irrigation furrows, burners are fre- < quently mounted on trailing skids which control the height of the burners and follow the land irregularities. Flame-cultivator burners Significant developments in burner de- sign and positioning during the years 1948 to 1953 were largely responsible for revival of interest in flame cultivation in the mid-South and in California. The development of the flat Stoneville burner in Mississippi and its modifications in the Arkansas burner occurred during this period. The flat Stoneville burner has a flame opening % inch by 5% inches and uses a double orifice fan-spray nozzle for the fuel orifice. The Arkansas burner is a modification of this design, and incor- porates a 30-degree deflector intended to minimize heat movement into the cotton. The optimum position of the two burners has been determined through ' coordinated laboratory and field trials. The Stoneville burner should be set at an angle between 30 and 45 degrees from p the horizontal. On smooth, well-formed beds, positioning the burner at the 30- degree burner angle will give better flame coverage over the bed. If beds are vari- able or rough, increasing the angle to approximately 45 degrees will cause less flame to be deflected into the cotton. The Arkansas burner should be ad- justed so that the combustion chamber is level. Then, because of its 30-degree flame deflector, its row coverage is iden- tical with that of the Stoneville burner set at 30 degrees. [12] Flame cultivator burners. Stoneville burner (left) and Arkansas burner. For normal flame cultivation a stand- ard Spraying Systems T-Jet 2-2502 noz- zle tip is used. This twin orifice, 25° angle, flat fan nozzle-tip delivers 0.20 gallons per minute at 40 psi and has been found to be the best for most conditions. When the 2-2502 nozzle tip is used, the distance from the tip of either type of burner to the point where the flame touches the bed should be 8 inches. In this position, the flame should strike the bed 2 inches on the burner side of the plant row. A steel rule laid along the top of the burner or deflector will aid in making these adjustments, as the ac- companying drawing indicates. Either burner will vary slightly in its optimum setting, according to the shape and slope of the row. The best setting for any burner-nozzle combination can be determined at night when the flame pattern is visible. The outlet end of the burners should be kept parallel to the row and level in order to insure uniform application of the flame, and the two burners, one on each side of the row, should be set in tandem so that flames will not oppose each other. No differences in weed control or cot- ton plant damage are found between the two types of burners when operated with the same adjustments. Minimum Cotton Height 6 to 8 Inches Stoneville Proper adjustment of burners in relation to cotton plant and bed profile. [13] Tractor speed and fuel pressure The speed of the flame-cultivator trac- tor is a critical factor if maximum weed killing and minimum cotton-plant dam- age are to be obtained. Tractor speeds less than 2% niph have resulted in mod- erate to severe damage to cotton plants, while speeds greater than 3% niph have reduced the effectiveness of flame for killing weeds. The most satisfactory re- sults have been obtained at speeds be- tween 2% and 3% mph. Within certain limits, the fuel pres- sure or rate of fuel consumption has less effect than tractor speed on damage to cotton. In general, as cotton height in- creases from 8 to 18 inches, fuel pres- sure may be increased from 35 to 50 psi for better weed control. Cost of flame cultivation Since fuel is the most costly single item in operating a flame cultivator, the cost of operation will vary with the price of LP gas, which in turn varies with quan- tity purchased and distance from source of supply. Cost of operation can be esti- mated by applying the local price of LP gas to the fuel consumption rates given in table 4 and adding tractor costs. Table 5 shows typical costs for mid-season flame cultivation. Table 4. Fuel Consumption With Different- Sized Nozzle Tips Used at Various Pressures and Using Two Burners per 40 " Row.* Orifice tip no. Gallons per acref 2-25015 80 psi 2.50 3.10 4.66 40 psi 3.18 3.84 5.73 50 psi 3.65 4.60 6.76 60 psi 4.26 2-2502| 5.30 2-2503 7.75 * Fuel consumption for vapor burners with a tractor cooling system temperature of 180°F. For liquid burners- rates will be approximately 10 to 20 per cent less, depending upon construction of heat exchanger. t No allowance made for turning. t Recommended nozzle tip. Source: Carter, L. M., R. F. Colwick, and J. R. Tavernetti. 1960. Evaluating flame-burner design for weed con- trol in cotton. Transactions of ASAE 3(2):125-28. Table 5. Typical Costs of Mid-Season Flame Cultivation Item Amount per acre Unit cost Total cost per acre Fuel 4£ gal i hour j hour dollars 0.12 1.25 1.83* dollars 54 Labor , 31 Tractor 46 Total. . 1.31 * Source: Reed, A. D. 1959. Machinery cost and related cost data. University of California Ext. Ser. (Mimeograph Report.) [14 CHEMICAL CONTROL OF ANNUAL WEEDS Sound farming practices are essential to successful weed control, and herbicides for weed control should be considered as supplements to, rather than replacements for, such practices. Chemical weed con- trol (if it can be used) offers a means of reducing labor costs in cotton produc- tion and may in some cases result in improved quality and/or increased yields. Pre-emergence weed control The weed control obtained in Cali- fornia with surface-applied pre-emer- gence herbicides in cotton has been erratic and unpredictable. To provide weed control, surface-applied herbicides must be water-borne from soil surface to the root zones of germinating weed seeds ; rain, or sprinkler irrigation, is most effec- tive for moving the herbicides, but furrow irrigation immediately after planting has not been successful. Rainfall in cotton-producing areas of California is extremely unpredictable at the time of planting. Table 6 gives the probabilities of various amounts of rain- fall from March 25 through April, at Madera, California (one of the cotton areas most likely to receive rainfall) . When rainfall of % inch or more oc- curred soon after herbicide application, certain herbicides controlled annual weeds satisfactorily. When rainfall was delayed, or did not occur after planting, pre-emergence herbicides failed to con- trol annual weeds adequately. Environmental conditions favorable to pre-emergence weed control in cotton are seldom found in California. The use of pre-emergence herbicides for annual weed control is thus limited to isolated areas where rainfall can be anticipated, or where sprinkler irrigation of % to 1 inch of water can be used. In these areas, pre-emergence herbicides may be applied as 10 to 12-inch bands over the drill row during or immediately after planting. The treated bands should be left undisturbed for maximum weed control. High-bed planting and planting to a stand (no thinning) will help to keep soil disturbance in the treated band to a minimum. When rain has occurred, CIPC has generally proved to be the best pre- emergence herbicide used in California although it is not very effective for an- nual broad-leaved weeds. Cost of CIPC at rates of 6 to 8 pounds per acre, ap- Table 6. Probability of Various Amounts of Rain in any Given Week During the Cotton-Planting Season at Madera, California (Based on a 34- Year Precipitation record)* Rainfall Probability of receiving stated amounts of rainfall March April (4th wk.) (1st wk.) (2nd wk.) (3rd wk.) (4th wk.) inches per cent 0.01+.. . 21.2 11.8 8.2 4.5 16.2 9.3 5.7 3.3 9.0 6.2 3.2 1.2 6.1 4.9 1.6 O.S 16.3 1+ 10 2 3+ 5 7 5+ 1 6 *Unpublished data. [15] plied in 10-inch bands, averages S3 to $4 per acre. Evidence that certain compounds are herbicidally active when incorporated into moist soil suggests that spraying the surface and mixing the herbicide into the soil, or using subsurface applications of herbicides, may be a logical approach to early season weed control. Preliminary research data likewise indicate the pos- sible advisability of applying herbicides before both pre-irrigation and planting. Should either of these practices prove effective, early weed control could be accomplished without reliance upon rain- fall. Early post-emergence weed control There are no reliable chemical pro- grams at present, except selective oiling, which provide weed control from the time of cotton emergence until a layby herbicide can be applied. Growers must rely upon mechanical means or upon directed oil applications until cultiva- tion can be supplemented with flame. Layby weed control If cotton produces large amounts of foliage, cultivation alone or cultivation supplemented by flame may be adequate to protect it from weeds for the remain- der of the season. However, cotton fields t which have been relatively free of weeds until layby may present difficult weed problems by harvest time because of late < weed emergence. This is especially true where the cotton stand is poor, or where vegetative growth is limited because of soil or other environmental conditions. In these cases, use of a herbicide imme- diately after the last cultivation can pro- vide weed control for the remainder of " the season. Layby herbicides generally must be applied before weeds emerge, and there- * fore the fields must be clean at time of application. Layby applications can be made with granular materials, or with £ broadcast-directed sprays in which the nozzle or nozzles are suspended midway between two rows of cotton. The spray pattern is adjusted to a height that will allow the pattern to overlap at the base Herbicidal spraying at layby of cotton. [16] of the cotton plant and yet spray a mini- mum of herbicide on the cotton foliage. Cotton plants are not injured when a few lower leaves are covered with the herbicide spray because these mature leaves manufacture little food for the plant after layby time, and thus their loss does not affect plant development. Shields or plant guards are advisable for partially lodged cotton, or for cotton which has heavily-fruited lower branches. The shields will permit more uniform cover- age of the soil, and less herbicide will be wasted on cotoon foliage. Layby herbicide applications should be followed by an irrigation soon after treatment. Control of the irrigation water to insure thorough wetting of the beds is essential to obtain successful weed con- trol. Table 7 shows the results of a 4-year study of the influence of layby weed con- trol with herbicides upon vigorous uni- form stands of cotton at the U. S. Cotton Field Station at Shafter, California. Arrangement of spraying nozzles between two rows of cotton. Table 7. Layby Weed Control With Herbicides in Cotton (4 Year Period) ; U. S. Cotton Field Station, Shafter, California Herbicide Rate of application Early layby weed rating* Layby weed ratingf 1955 1956 1957 1958 1955 1956 1957 1958 pounds per acre 0.0 6.0 9.0 6.0 9.0 1.0 1.5 1.0 1.5 6.0 9.0 2.6 6.8 6.6 8.6 8.5 7.3 8.8 4.1 6.2 5.7 6.9 7.4 8.5 8.9 7.0 8.0 6.8 7.4 4.5 7.4 7.7 7.5 7.1 8.3 7.8 8.6 5.2 4.6 6.9 7.8 8.4 9.1 8.8 9.0 9.1 5.9 9.0 9.1 6.9 7.4 9.0 9.1 6.7 7.2 7.5 7.8 8.3 8.3 9.0 8.7 8.0 7.8 7.9 6.2 9.1 8.0 6.7 7.7 8.9 8.6 9.4 6.2 7.3 8.0 NPA 9.1 CIPC 8.4 8.8 Monuron 9.0 9.6 9.5 9 6 CDAA 9.6 8.5 8.6 Statistical analysis LSD .05 LSD .01 0.9 1.2 1.6 2.0 1.6 2.2 1 3 1.8 2.4 N.S. N.S. N.S. 1.8 2.4 1.0 N.S. * Early layby: cotton 12-24* in height; weed ratings: = no control; 10 = complete control, t Layby: cotton 24-48" in height; weed ratings: = no control; 10 = complete control. Source: Miller, J. H., H. M. Kempen, J. A. Wilkerson, and C. L. Foy, Control of barnyardgrass {Echinochloa crusgalli) in western irrigated cotton. Weeds. 9(2):273-81. [17] Effect of different irrigations following herbicides applied at layby. (Top) Diuron applied at 1.6 pounds per acre / followed by standard irrigation. Note grass in drill row where water did not advance across top of bed. Foreground grass is dead. (Bottom) Diuron applied at same rate as above / followed by irrigation which completely wetted top of beds. Note absence of grass in drill row. A comparison of herbicides applied at layby shows that the most effective weed control was obtained with diuron and monuron.* However, advantages of a herbicide application after cotton was 24 to 48 inches high were few because of the vigorous competitiveness of the cot- ton crop with late weeds. Where cotton stands are poorer, or where cotton * Diuron is registered for use in cotton in all of California; monuron is registered for use south of the Tehachapi Mountains only. growth is limited, advantages of layby herbicides are much more pronounced. In the latter instances, selective herbicides having residual activity (those that re- main active in the soil for 2 months or more) have been most effective. Diuron and monuron fall into this class, whereas CIPC, which remains active in the soil for not over 4 to 6 weeks under these conditions, is intermediate. Thus diuron and monuron generally provide excellent weed control after a layby application. [18] CIPC is effective on annual grasses but is relatively ineffective on broad-leaved weeds. Diuron and monuron are wettable powders, and are only slightly soluble in water. Because of this, vigorous agi- tation must be maintained in the sprayer tank to keep the herbicides in suspen- sion; mechanical agitation is more posi- tive and is preferred to the bypass method of agitation. Both herbicides have a tendency to clog screens in the sprayer lines and screens smaller than 50 mesh should not be used. Extreme caution should be exercised in calculating rates of herbicides. Proper calibration of spray equipment likewise cannot be over-emphasized (see "Appen- dix"). The suggested rates of applica- tion of monuron range from 1 to 1% pounds per acre of 80 per cent material, depending upon soil; diuron should be used at rates of 1 to 2 pounds per acre of 80 per cent material. Diuron and mon- uron are adsorbed on the clay fraction of the soil, so that in soils containing large amounts of clay the higher rates quoted above are necessary for weed control. Neither of these herbicides is recommended for use on sands. Because of the influence of soil prop- erties on diuron and monuron, growers should at first try different rates of ap- plication on a limited acreage only. Where cotton cannot be followed by cot- ton, corn, or sorghum, CIPC may be used. CIPC should be used at the rate of 8 pounds per acre (on an active in- gredient basis) . The granular formula- tion provides more satisfactory weed control than the emulsified formulations. Water must be applied soon after the CIPC application. Time of application. A layby ap- plication of monuron, diuron, or CIPC made during the last half of June will provide weed control for the remainder of the season without injury to cotton plants. Applications made at this time do not require high-clearance equipment, shields, or plant guards, and the number of sweep cultivations is reduced. In areas where there is difficulty with water penetration, however, fewer sweep appli- cations may increase the problem. Cotton plant injury may result if her- bicides are applied too early, but usually results only when a large percentage of the total leaf area of the cotton plants is covered with spray. No cotton injury from directed spray has been encoun- tered when plants were 12 inches tall, or higher. The proper use of monuron, diuron, or CIPC has not reduced cotton yields and has improved grades of lint cotton mate- rially, as compared to normally cultivated cotton where weeds are present — the im- proved grades have resulted from reduc- tion of trash in the lint cotton. Cotton fiber qualities have not been altered by these herbicides. Most annual broad-leaved and grass weeds are controlled by diuron or mon- uron; these include common morning- glory, rough pigweed, common lambs- quarters, ground-cherry, purslane, barn- yardgrass (watergrass), witchgrass, stinkgrass, and red sprangletop. Weeds not satisfactorily controlled include punc- turevine, field bindweed (perennial morning-glory), established Johnson- grass and Bermudagrass, and nutsedge (nutgrass) . CIPC will control annual grasses but is less effective on broad- leaved weeds. Residual effects of layby herbi- cides. Layby applications of diuron and monuron may be active in the soil after the cotton crop is harvested, and all winter-planted crops and many spring- For current recommendations and new developments in the use of selective herbicides in cotton and in other crops consult the latest "University of Califor- nia Weed Control Recommendations," obtainable from your local Farm Ad- visor. [19 Because of the confusion often encountered in the use of common names, the Weed Society of America (WSA) has attempted to standardize terminology in the weed control field. Weeds and chemicals mentioned in this bulletin are listed below by their common and scientific names. Weeds WSA common name Other common name (s) Scientific name field bindweed wild mustard puncturevine common morning-glory pigweed lambs-quarters ground cherry (not listed) common purslane barnyardgrass witchgrass stinkgrass sprangletop (not listed) Johnsongrass Bermudagrass yellow nutsedge Russian knapweed large crabgrass silverleaf night shade wild morning-glory field mustard bull head, goat head garden morning-glory red root, careless weed fat hen, white pigweed tomatillo pussley, pursley watergrass, cocksfoot ticklegrass snakegrass, lovegrass scale grass, ray grass Means grass devilgrass, wire grass yellow nutgrass, ground almond Turkestan thistle hairy crabgrass white horsenettle Convolvulus arvensis Brassica kaber var. pinnatifida Tribulus terrestris Ipomoea purpurea Amaranthus retro flexus Chenopodium album Physalis ixocarpa Portulaca oleracea Echinochloa crusgalli Panicum capillare Eragrostis cilianensis Leptochloa fascicularis Sorgum halepense Cynodon dactylon Cyperus esculentus Centaurea repens Digitaria sanguinalis Solatium elaeagnifolium Chemicals WSA common name Other common, commercial, or trade names Chemical name CIPC NPA CDAA diuron monuron TCA dalapon HCA SMDC PBA 2,3,6-TBA fenuron simazine Chloro IPC Alanap Randox Karmex Telvar Sodium TCA Dowpon Vapam, VPM Benzac 354 Trysben 200, Benzac 1281 Dybar Simazine isopropyl N-(3-chlorophenyl) carbamate N-1-naphthylphthalamic acid 2-chloro-N,N-diallylacetamide 3- (3,4-dichlorophenyl) -1,1-dimethylurea 3- (p-chlorophenyl) -1,1-dimethylurea trichloroacetic acid 2,2-dichloropropionic acid hexachloroacetone sodium-methyl-dithiocarbamate polychlorobenzoic acid 2,3,6-trichlorobenzoic acid 3-phenyl-l,l-dimethylurea 2-chloro-4,6-bis (ethylamino) -s-triazine [20] planted ones could be injured by these residues. Therefore, only crops known to be tolerant to small amounts of diuron or monuron (such as cotton, corn, or sor- ghum) should be planted the season after a layby application of these materials; deep moldboard plowing after harvest will reduce toxicity of residues, but not sufficiently to permit safe cropping of nontolerant crops. CIPC will not remain in the soil to injure winter-planted crops, and chemical breakdown of soil-applied diuron or monuron is virtually complete in 12 to 15 months. No residual accumu- lation was detected over a 5-year period in which monuron and diuron were ap- plied annually at layby time on a Hes- peria fine sandy loam. BIOLOGICAL METHODS OF WEED CONTROL Geese have given excellent results in controlling grasses in cotton as they will feed on Johnsongrass, Bermudagrass, and nutsedge and seedlings of barnyard- grass, large crabgrass, witchgrass, and most other grasses, with little or no feed- ing on cotton or broad-leaved weeds. Geese should be present in the field from the time grasses emerge. Several management problems must be considered when using geese. Clean drinking water must be available at all times but must be moved with regularity to train the birds to range; one or two drinking troughs in 20 acres filled daily should be sufficient. Supplemental feed must be used sparingly, as geese will not feed on grass if plentiful grain feeds are available. Pelleted feeds for poultry or rabbit are satisfactory and should be used at rates of about 10 to 20 pounds daily per 100 birds. Feeding areas should be moved regularly. Old geese cannot be trained to feed on grass, and young goslings must be used each season. They should be at least 6 weeks of age and well-feathered before being placed in the fields. Maintaining the proper balance be- tween numbers of geese and weed popu- lations is a major problem. If too few geese are used, they will be unable to control the growth of the weeds but too many geese for the weed population will require more supplemental feeding of the birds. Grass alone is not a well- balanced diet for geese and some deaths may occur because of compaction of the digestive tract. Normally, 3 to 5 geese per acre are sufficient to control most grass populations in cotton but after the first year the number can usually be re- duced to 1 or 2 per acre; this number will vary with the density of the grass. Shade must be provided for geese dur- ing the highest temperatures of the day during the warmest part of the season, and where natural shade is lacking cot- ton trailers are often used. Unless the shades are portable and are moved with regularity, the geese will damage cotton immediately adjacent to the shade. Geese range more effectively during the cooler portions of the day, and are even more effective if allowed to range at night — although bird deaths due to predators is greatest then; dogs are by far the worst offenders in destroying geese. A 2 or 3-foot high 2-inch mesh poultry fence supported with light stakes or lath is generally used to confine young geese, but will not exclude predatory ani- mals. Whenever they cannot be watched closely at night, geese should be protected. When toxic chemicals are used in the area, geese should be penned, as many chemicals are dangerous to the birds. After the weeding season, the geese [21 can be sold or returned to the hatchery for credit. Most dealers contract to buy the mature geese for about 1 dollar per bird. Cost of weeding with geese will vary considerably from farm to farm. The number of geese, fencing, feed and water facilities, and extra labor are the major items. The following is an estimate of annual costs for a 40-acre field: 120 geese at $2 (less SI credit for half of the geese) $180 Wire and stakes $150 (pro-rated over 3 years) $ 50 Grain $ 60 Water and watering equipment $ 80 Labor and incidentals $150 Total $520 Annual cost per acre $ 13 CONTROL OF PERENNIAL WEEDS Johnsongrass and Bermudagrass Cultural methods. Dense stands of Johnsongrass or Bermudagrass in cotton cannot be economically controlled with herbicides at present, because the herbi- cides that control them also injure cot- ton. Where dense stands of such grasses occur, nonselective methods, such as sum- mer fallowing or spraying with nonselec- tive herbicides, should be used to reduce them. Fallowing may be used to reduce grass population to scattered plants before herbicides are used as clean-up treat- ments in cotton. Following early-matur- ing crops, such as small grain or pota- toes, the land may be fallowed for the remainder of the season. Cultivation with such equipment as the spring-tooth har- row after plowing will bring the rhi- zomes (see "Definition of Terms," page 4) to the surface of the soil, and ex- posure of the rhizomes to the hot, drying sun is an effective means of destroying much of the grass. Cultivations should be repeated at frequent intervals during the summer — for example, Johnsongrass growth should not exceed 12 inches be- tween operations. The use of nonselective herbicides as broadcast sprays may be advisable for isolated, densely populated patches of perennial grass in fields. Soil sterilants. Winter applications of the sodium salt of trichloroacetic acid (TCA) at rates of 100 to 150 pounds per acre have in some cases almost eradi- , cated perennial grasses. Rainfall, or irri- gation, sufficient to carry the herbicide into the root zone of the grass is neces- sary, but too much water may carry the TCA below the root zone of the grass. Much TCA can be lost if exposed to high temperatures before rainfall or irriga- tion occurs and, in general, the rather exacting conditions necessary for good performance make it difficult to predict the effectiveness of TCA. Residues of this chemical in the soil may occasionally be £ toxic to spring crops following treatment, although tillage of the soil, and irriga- tion, will usually reduce residues to an insignificant amount. Flood irrigation is vastly superior to furrow irrigation for the removal of TCA. -,< Sodium chlorate may be used as a semi-permanent soil sterilant for the con- trol of Johnsongrass and Bermudagrass. This herbicide is also dependent upon water to carry it into the root zone of the plant. The suggested rate, 1% to 3 pounds per 100 square feet, will render the soil unproductive for 1 or more years, depending upon rainfall, or irrigation schedules. SODIUM CHLORATE IS A ' FIRE HAZARD IF COMBINED WITH FLAMMABLE MATERIALS SUCH AS CLOTHING OR PLANT RESIDUES < [22] X3 fl ctf bD Pi a> o ffi ^ T3 10 >-l tH d ffi CO >> o ^2 •i-H T3 0 O] OS OS CO CO CO CI ~r CI OS id OS CO O t~ o CO ^ •V CO ■>* CO (^ CO o >o CO CO 00 — i •& O o a> ^ •S CO T3 8 OT3 $£ t * £ S 5 .5 J[ ■U^^ O e e 0) e ,C , — • -i^> co co co § .2 2 -g .2 .2 .2 KtftcuQQQ 3 3 O O P P CO CO OS OS co O fa a . CO 5 £ p p /. X O -o cp o Oco 3Ch c c ii ■S&i CO q mile; therefore, MPH = 180 Sec. to go 264 feet.) 2. Start with a full tank of clean water. 3. Determine the number of gallons used in 6 minutes by one of the following methods (use a stop watch or a watch with a second hand to time the 6 minutes and measure the water to the nearest pint) : a) Place a bucket under each nozzle and measure the total amount of water sprayed in 6 minutes, or b) Measure the amount of water required to refill tank after the 6-minute run. 4. Calculate spray coverage in inches. (See "Notes on calibration," next page.) 5. Substitute this information in the following formula* : 1000 x gallons used in 6 minutes Gallons per acre = ■ width of spray coverage in inches x miles per hour. 6. If this is more than the amount desired, decrease pressure; if less, increase pres- sure and recalibrate. Method 2. Procedure: 1. Start with a full tank of clean water. 2. Operate for % mile (1320 feet) with the determined pressure and with the tractor throttle in the marked operating position. 3. Refill tank and convert the amount to gallons and fractions of gallons (nearest pint). 4. Calculate spray coverage in inches. (See "Notes on calibration," next page.) 5. Substitute this information in the following formulaf : 400 x gallons used in % mile Gallons per acre = width of spray coverage in inches. 6. If this figure is more than the amount desired, decrease pressure; if less, increase pressure and recalibrate. * Accurate to 1%. If greater accuracy is desired, substitute the factor 990 for the factor 1000 in this formula. t Accurate to 1%. If greater accuracy is desired, substitute the factor 396 for the factor 400 in this formula. [29] Notes on calibration 1. Boom lines must be filled before beginning. 2. Nozzles must be delivering uniform amounts (within 10 per cent) . 3. Speed or pressure must not be changed during course of the calibration. 4. Accurate measurement of the amount of water sprayed is essential. Tank should be level when refilling to the pre-determined mark. Error should be less than 10 per cent in duplicate trials. 5. The width of spray coverage for band application is the sum of the band widths ; for 10-inch bands and a 4-row spray boom the spray coverage is 40 inches. In broadcast application the spray width is the effective boom coverage — for ex- ample, for 40-inch rows and 4-row spray coverage the width is 160 inches. Use the following method to determine the amount of herbicide to be put in the sprayer tank: 1. To obtain the number of acres one filling will spray, divide the number of gal- lons the tank will hold by number of gallons applied per acre. 2. To obtain the amount to be used per tank, multiply number of acres the tank will spray by the amount of herbicide to be used per acre. [30 In order that the information in our publications may be more intelligible it is sometimes necessary to use trade names of products or equipment rather than complicated descriptive or chemical iden- tifications. In so doing it is unavoidable in some cases that similar products which are on the market under other trade names may not be cited. No endorsement of named products is intended nor is criticism implied of similar products which are not mentioned. 25m-12,'62(D2404)VL How to do it . . .in photos New (and sometimes old) techniques are described and illustrated for better understanding of ', at times, complicated subjects. The rule is, "If it can't be described, use a photo; if a photo won't do, draw a picture." :*&&;. W.;- Some Systems Work; Some Don't Scientists at the University of California are constantly trying new plant varie- ties, new growing techniques, new machinery, in an effort to improve the State's agriculture. Their findings are reported and, when possible, illustrated in tech- nical, semi-technical, and popular publications that are available to anyone. Perhaps the answer to your farming problem is in one or more of these pub- lications. For a catalog, write to: AGRICULTURAL PUBLICATIONS University of California • 207 University Hall Berkeley 4 *'