Two separate variety evaluations were conducted in southeastern Arizona during the 2004 cotton growing season. The two locations were on grower-cooperator fields in the Upper Gila River Valley located in Thatcher, AZ (Graham County) and in the Sulfur Springs Valley in Kansas Settlement, AZ (Cochise County). Twelve varieties were selected for the Graham County evaluation and fifteen in the Cochise County evaluation. These varieties included several transgenic varieties and ranged in maturity from early to full-season varieties. Several Acala varieties were also evaluated in both the Graham and Cochise County tests. Both evaluations were conducted using a randomized complete block design with each variety replicated four times. Plant measurements were collected in season on several dates from the Graham County evaluation. End of season plant measurements were collected from the Cochise County evaluation. Lint yield was estimated at each location by harvesting the entire plot and weighing the harvested seed cotton with a weigh wagon equipped with load cells. Sub samples were collected from each plot for fiber quality and percent lint determinations. Total crop value for each variety was calculated by using the fiber quality premium/discount and using a $0.52 per pound price. The total price is then multiplied by total lint yield to obtain the total value for that particular variety. Results observed in the Graham County evaluation were similar to those in 2003. Lint yield ranged from 1200 to over 1600 lbs. lint/acre. The FiberMax variety FM991BR produced the highest lint yield and also the highest total crop value at over $950/acre. Results from the Cochise County evaluation demonstrated the potential that high fiber quality can have on total crop value. Lint yields ranged from 600 to over 1200 lbs. lint/acre. The highest yielding variety (ST5242BR) did not produce the highest crop value. Because of the higher fiber quality of the Acala varieties, they produced the highest value at approximately $630/acre.

Each year the University of Arizona conducts variety trials across the state to evaluate the performance of upland cotton varieties. These tests provide unbiased data on the performance of varieties when tested side-by-side under typical production practices. In 2004 we planted a total of 11 trials, two in the Yuma region (Yuma County), two in the western region (LaPaz and Mohave counties), four in the central region (Maricopa and Pinal counties), one in the southern region (Pima county), and two in the eastern region (Graham and Cochise counties). We tested seven to eight commercially available varieties at each test site.

The tillage operations required to grow an annual barley and cotton crop rotation were reduced by eliminating tillage prior to planting cotton, eliminating cultivations for weed control in cotton, and especially by eliminating tillage following cotton. A light activated, weed sensing automatic spot-spray system reduced the amount of spray volume and herbicide used by 40% to 60% at Marana and 36% to 56% at Maricopa in 2004. At Maricopa, a large number of volunteer cotton plants in the furrows of early planted no-till cotton reduced the spray volume savings from using the weed sensing automatic spot-spray system. Weed control was similar with the weed sensing, automatic spot-spray system compared to the conventional continuous spray system for most weed species but weeds with narrow leaf, upright leaf canopies such as sprangletop, barley and skeleton weed were more difficult to detect and control. In both Marana and Maricopa, there were yield differences between treatments related to planting date, with late-planted cotton yielding less than early-planted cotton. At Marana, the early-planted conventional tillage cotton out-yielded the barley cover crop, early-planted no-till cotton treatment. At Maricopa, there were no yield differences between the two early planted cotton treatments; however, the late-planted conventionally tilled cotton yielded 28% more than the late-planted no-till cotton. Although the yield comparisons are not yet definitive, it appears that in some situations no-till cotton may yield less than conventionally tilled cotton. At Maricopa, the height of cereal crop stubble did not affect subsequent cotton establishment, field populations, plant height or lint production (2003 and 2004) and the position or node of the first fruiting branch and the first retained boll were similarly unaffected in 2004.

A field experiment was conducted in 2004 at the University of Arizona Maricopa Agricultural Center (1,175ft. elevation) to evaluate the effects of five irrigation termination (IT1, IT2, IT3, IT4, and IT5) dates on yield and fiber micronaire of several Upland cotton varieties. In addition, the economic relationships of IT treatments were also evaluated. The first IT treatment (IT1) was made with the intention of terminating irrigations somewhat pre-maturely. Based upon current UA recommendations for IT to complete a single cycle fruit set, the more optimal date of IT would have included one or two additional irrigations (beyond IT1). In this experiment, IT2 was structured to provide an additional (one) irrigation before the more optimal date. For the IT3 plots, the intention was to attempt to time termination to match the conventional growers optimal date. The IT4 and IT5 were imposed to attempt to produce a second cycle fruit set and irrigations continued until 27 August and 21 September respectively. In general, lint yield and micronaire results revealed significant differences among the IT treatments. In a similar fashion to 2000-2002 IT experiments, micronaire and lint yield values consistently increased with later IT dates. The best micronaire and lint yield results were achieved with IT4 date, which received 12 in. less irrigation water than IT5. The 12 in. water saved equates to approximately 20% of the total water used under the conventional practice. The average marginal value of water for all Upland varieties in going from IT1 to IT2, IT2 to IT3, IT3 to IT4, and IT4 to IT5 using November 2004 prices and low carrying costs is calculated at $320.07, $150.15, $100.54, and -$28.16 per acre-foot of water. If steeper mike discounts (November 1999), a lower base lint price (45¢/lb.), and higher costs (i.e., more costly insecticide and chemical costs) are imputed to extend the crop, the marginal value of an acre-foot of water for all Upland varieties and replications in going from IT1 to IT2, IT2 to IT3, IT3 to IT4, and IT4 to IT5 is estimated at $164.04, $48.15, $12.97, and -$94.79. Profitability and marginal value of water sometimes vary quite markedly between different varieties and termination dates as well.

Two experiments were conducted in 2004 evaluating twin line cotton production using a conservation tillage system approach. DPL 451 BR Upland cotton was planted into oat hay stubble on April 30 and May 5, 2004 at commercial cooperator sites at Tonopah and Tolleson AZ, respectively. The two primary experimental objectives were to determine whether cotton planted into previous crop residue initiated fruiting on the mainstem once the cotton seedling grew above the crop stubble and whether there were differences in lint yield between the single and twin line system. Previous twin line cotton production research had been conducted by authors at 30 locations from 2001-2003. In almost all cases, the harvest of low set bolls presented problems with the twin line system. In 2004, the initiation of the first fruiting branch was independent of the stubble height at both locations. In addition, there were no differences in lint yield in either a single or twin line cotton production system when planting into previous crop residue using conservation tillage.

A single field trial was conducted during the 2004 season to evaluate three different mepiquat formulations used to control vegetative plant growth. This project involved the evaluation of mepiquat formulations of Mepex Ginout (DuPont), Pix Ultra and Pentia (BASF). These three treatments, along with a control, were arranged in a randomized complete block design with four replications. Plots were 20, 36” wide rows and extended for a full one half mile irrigation run. Applications of all treatments were made on 22 July 2004 at a rate of 16 oz/acre for each formulation. Lint yields were determined at the end of the season by harvesting the center eight rows of each plot. The harvested seed cotton was weighed using a weigh wagon equipped with load cells. Sub samples were collected from each plot for fiber quality and lint percent determinations. No significant differences were observed among treatments with respect to any of the fiber quality parameters measured. The Mepex Ginout treatment produced the highest lint yield and also the highest percent lint. The control was not significantly different from the other two formulation treatments. Lint percent values had a major impact on lint yield. The control had the lowest seed cotton yield but because of the higher percent lint values, lint yield for the control was ultimately higher than both the Pentia and Pix Ultra treatments. This study demonstrates the positive effect of the PGR applications and also the effect that percent lint can have on final lint yield.

A single defoliation experiment was conducted during the 2004 growing season in an effort to evaluate the effectiveness of commercially available harvest preparation materials in the higher elevation areas of southeastern Arizona. Many growers will not employ a defoliation regime but will let the crop naturally senesce with the cooler temperatures and frost. This evaluation compared a standard treatment of Ginstar with some additional tank mix ethephon based products from DuPont and BASF. These treatments were compared to the common Na chlorate treatment employed by many growers in this region of the state. Treatments included a base rate of 8 oz/acre Ginstar with three different rates of both CottonQuick and Prep. Treatments were applied on 15 October in a randomized complete block design with four replications. Data collected included observations of percent defoliation, percent regrowth control, and percent open boll on two separate dates after treatment (27 October and 4 November). Yield data was also collected at the end of the season by harvesting the center two rows of each plot. Sub samples were collected for fiber quality analysis. Results showed significant differences among treatments with respect to defoliation parameters measured. Treatments including Ginstar and higher rates of both CottonQuick and Prep performed well. The treatment consisting of Na chlorate alone was also effective. The Ginstar + CottonQuick treatment did appear to have a slight advantage in final percent open boll counts. No significant differences among treatments were detected with respect to yield and fiber quality. However, lint yield for the control treatment was lower than the other defoliation treatments. This would indicate that some type of defoliation regime does provide benefit in terms of increased yield.

Upland cotton advanced strains and commercial check varieties were evaluated in replicated field studies at three locations in 2004. The test sites include the AZ locations of Safford, Maricopa, and Yuma. Eight seed companies submitted a maximum of ten advanced strain entries per location. Three commercial check varieties were used at all three sites, and included ST5599BR, DP449BR, and DP448B. Data collected included final plant heights, yield, and fiber quality. The research is conducted in order to develop public unbiased performance data of genetic materials that have moved to the advanced stages of testing and are being considered for commercial release. The data have historically been used to add to seed company databases and assist with commercial release decisions.

Lygus hesperus (Knight) has been the number one pest of Arizona cotton since 1998. With advances in the selective management of Arizona cotton’s other two key pests (i.e., Bemisia tabaci and Pectinophora gossypiella), there has been less opportunity for collateral control of this yield-limiting pest. There has also been a new premium placed on locating less disruptive, even selective, control methods that are compatible with existing selective technologies. Our laboratory routinely screens candidate compounds for efficacy against Lygus hesperus under desert conditions. Promising leads are further developed and evaluated in the field for efficacy, spectrum of activity and safety for beneficial arthropods. Our recent findings have identified three compounds with potential for delivering economic control of Lygus hesperus with greater safety for beneficial arthropods than current standards of control. At the same time, our work has importantly identified many compounds that are ineffective against our Lygus, despite reported success against a related species, the tarnished plant bug, Lygus lineolaris (e.g., acetamiprid, imidacloprid, indoxacarb, pyrethroids, thiamethoxam). Flonicamid, a pyridine carboxamide, is under development by FMC in the U.S. This aphicide has shown excellent results in controlling our Lygus in cotton. Novaluron, a chitin inhibitor under development by Uniroyal (as Diamond®) in the U.S., has rate sensitive activity against Lygus in cotton. Metaflumizone (BAS320I), under development by BASF in the U.S., is a semi-carbazone chemistry with significant efficacy against Lygus hesperus. These three or potentially other new leads in insecticide discovery may play an important role in transitioning Arizona cotton growers away from neuro-toxic, broad-spectrum, and disruptive organophosphates and carbamates currently used to control Lygus in cotton. The potential benefits to natural enemy conservation should help stabilize insect pest management in Arizona cotton, thus minimizing the chances of secondary pest outbreaks and costly pest resurgences. Until selective alternatives are found and registered, acephate (e.g., Orthene 97 by Valent USA) and oxamyl (Vydate C-LV by DuPont) remain our standard recommendations for Lygus control in Arizona cotton.

A set of experiments were conducted in two Arizona locations to evaluate preliminary and advanced strains from the Arizona Cotton Growers Association breeding program. These trials were conducted in Yuma, AZ (130 ft. above MSL) and Maricopa, AZ (1170 ft. above MSL). Strains were planted into four row plots extending 38 feet in a randomized complete block design with a minimum of three replications. Each location had three commercial cotton varieties included as control treatments for comparison. Data collected on these trials included a series of plant measurements at three growth stages over the course of the season, plus yield and fiber quality data. All data were subjected to statistical analysis to test for differences among strains for yield and fiber quality. The trial conducted at Yuma performed extremely well with yields ranging from 1300 to over 2100 lbs. lint/acre. The Maricopa test site experienced some adverse weather (high levels of heat stress) and insect related conditions that led to overall decreased yields. Ranges in yield were from approximately 800 to just over 1400 lbs. lint/acre. Significant differences among lint yield and fiber properties were experienced at each location. Plant measurement results also demonstrated interesting trends among strains and locations.