Field experiments were initiated in 2001 for a cotton-based conservation tillage project at sites in Marana, Coolidge, and Goodyear, Arizona. For the 2002 season, soil and water management assessments were made to evaluate surface irrigation performance under conservation tillage, following cover and grain crops. An additional site was added in the winter of 2002 at Maricopa, Arizona. Analyses included soil texture, infiltration rate, and water advancement. At Coolidge, conservation tillage plots had higher infiltration rates and longer advance times than the conventional plots in 2002 and 2003. At Marana, infiltration rates were initially higher for the Conservation plots but the rates converged at the end of four hours in 2002. In 2003, Conventional plots infiltrated about one inch more. Advance times for Marana showed water infiltration for Conventional wheel rows to be the fastest. At Goodyear, Conservation plots infiltrated more than Conventional plots during 2002. This also resulted in a slower advance time for the Conservation plots. In 2003, treatment effects were not comparable due to tillage by the grower. At Maricopa, Conservation plots infiltrated almost 2.2 inches more water than Conventional plots and water reached the end of the field three hours ahead of the fastest Conservation plot. Seasonal irrigation water applications to each treatment were relatively equal for all sites with the exception of Coolidge. Here, the long field combined with sandy soil made it difficult to adequately irrigate the Conservation plots. In 2002, an additional 21 inches of water was applied to the Conservation plots. In 2003, that amount was reduced to 12.5 inches. The yield data show a significant difference in 2002 only at Coolidge. There, Conservation plots yielded more than the Conventional ones. This may have been due to more water applied to the Conservation plots. In 2003, the opposite occurred with the Conventional plots yielding more than the Conservation plots. This could have been due to herbicide damage. At Maricopa, the Conventional plot also yielded more than the Conservation plot. Initial indications are that the conservation tillage does impact irrigation performance and may not be suitable for all locations depending on soil type and field layout.

Foliar fertilizers are not widely used for cotton production in the low desert, and data about their effects on cotton production under these conditions is therefore limited. This study documented the effects of Calcium Metalosate7 and Megafol, each applied at the rate of 1 qt/acre to DPL555BR cotton. Treatments were applied on July 7, and plants had been growing vigorously just prior to application. Plots were approximately 0.75 acres in size with four replications. Plant mapping data from late July indicated that non-treated cotton had numerically higher retention rates at each of the first three fruiting positions mapped in addition to slightly more total nodes and a greater number of reproductive nodes in part due to first fruiting structure being retained lower on the plant (node 6.75) than treated cotton (node 7.95 for Megafol, node 8.4 for Calcium Metalosate7). Tractor passage through treated plots may have also knocked off developing squares however. No statistical differences were noted for lbs. of lint/acre, although treated cotton did have slightly higher yields than the untreated check (1,162 lbs/acre) and treatments were almost identical (1,203 lbs./acre for Megafol; 1,198 lbs./acre for Calcium Metalosate7). Fiber lengths and strengths were significantly different by treatment, with shortest and weakest fibers resulting from cotton treated with Megafol. Cotton from Calcium Metalosate7 treatments were significantly longer and stronger than lint from Megafol treated cotton plots, but lint from untreated cotton plots was significantly longer and stronger than either treatment (36.6 staple, 31.0 g/tex). The reasons for these differences are unclear. It is difficult to correlate the slight yield increases noted with treatments in early July, especially in light of lower retention rates noted with treatments from plant mapping data in late July and the large amount of lint production that occurred in late 2003 due to summer heat. Multiple differences were noted for treatments in regards to lint quality, however, indicating treatments did affect cotton production. Size of bolls and cotton lint from these early summer bolls was not obtained but may have been an overlooked aspect of this study.

An experiment to investigate the effects of several plant growth enhancement products (AuxiGro7 WP, FirstChoice7 Bollster, CalMax, FoliGro7 BollSet) on DPL 451BR was conducted in the Palo Verde Valley. The cotton field chosen for this project was planted in April 2003, and flowering and boll/lint production was highly affected by summer heat, therefore product effectiveness was tested under the extreme high temperatures conditions that existed in 2003. Products were applied July 11 (shortly after first bloom), August 1st and/or August 15. Three rates (1, 2, 4 oz./acre) of AuxiGro7 WP were applied at each date, although just a single rate of other products was evaluated. Data collected included plant mapping following the first two application dates as well as lint yield and quality at harvest on October 20-23. Data indicated rate effects for treatments containing AuxiGro7 WP as noted in response to applications on August 1 (retention rates) and August 15 (cotton quality factors). These responses were signficantly different than the untreated check, and were often signficantly different than the Bollster fertilizer treatment itself as though Bollster was also included with AuxiGro7 WP treatments. Although yields for any treatment and the untreated check were not significantly different, previously noted differences for retention for the AuxiGro7 WP treatments may not have been realized due to the high temperatures and extended periods of not only Level 1 but Level 2 stress during the course of the experiment. Cotton values per acre were highest ($609.85/acre) in the 4 oz/acre of AuxiGro7 applied on August 15, due to highest valued lint as a result of micronaire of 5.2 combined with increased fiber strength and fiber length. Application of CalMax resulted in a significant increase in fiber strength following two applications of 4 oz./acre of AuxiGro7 + Bollster, but CalMax following two applications of Bollster did not affect fiber strength when compared with each of the two application treatments. A third application of FoliGro7 BollSet was detrimental for cotton yields and strength compared with only two applications of this product, although both treatment regimens resulted in shorter fibers that were not quite as strong as the untreated check.

Upland cotton advanced strains and commercial check varieties were evaluated in replicated field studies at three locations in 2003. The test sites include Safford, AZ, Maricopa, AZ, and Yuma, AZ. Seven seed companies submitted a maximum of ten advanced strains entries per location. Three commercial check varieties were used at the Safford and Yuma sites, and included ST4892BR, DP449BR, and DP565. Two commercial check varieties were used at the Maricopa site and included ST4892BR and DP449BR. Data collected included vigor and relative maturity ratings, 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 has historically been used to add to seed company databases and assist with commercial release decisions.

Eight Acala varieties from New Mexico(4), California(3) and Arizona(1) along with thirteen upland varieties of interest to the area were tested in a replicated small plot trial on the Safford Agricultural Center in Graham county at an elevation of 2950 feet. The highest yielding variety in this study was DP 555BR with a yield of 2203 pounds of lint per acre, but FM991BR produced the highest gross revenue per acre at $1527 per acre. Sierra, a new Acala variety from California, was the highest yielding Acala with a lint yield of 1872 pounds per acre and a gross revenue of $1324 per acre. The average lint yield was 500 pounds per acre higher that the 2002 season indicating that 2003 was an above average cotton growing year and that the heat units per day in the week following planting were above the threshold. In addition to the yield , several other agronomic variables were measured. These included plant height, total nodes, and boll weights. From plant height and total nodes the height to node ratios were calculated. Differences were seen between these variables by variety but the most notable point was that the plants were robust in their growth habit and fruiting forms were heavier than the previous year. HVI fiber quality data were reported and estimated values (in cents per pound of lint) were calculated. The HVI data showed an average fiber length of 1.13 inches, with only one variety producing a less than 1.10, and seven varieties having fiber of 1.15 inches or longer. A New Mexico experimental variety had the longest fiber at 1.18 inches. The average fiber strength was 32.1 grams per tex and the same NM experimental produced the strongest fiber. In general, all of the varieties included in this study had very good fiber.

Monitoring of Arizona pink bollworm (PBW), Pectinophora gossypiella, susceptibility to the Bt toxin Cry1Ac has been conducted annually since 1997. PBW were collected from cotton fields located throughout the Southwest in 2002, cultured in the laboratory, and tested for susceptibility to Cry1Ac using diet-incorporation bioassays. A total of 13 Arizona collections were successfully reared and bioassayed. Collections from California (6), New Mexico (1), and Texas (1) were also tested. Laboratory selection of pink bollworms collected from Arizona in 1997 and exposed to Cry1Ac in diet produced a strain capable of surviving on Bollgard® cotton. Subsequent studies showed that 10 g Cry1Ac/ml of insect diet was a reliable diagnostic concentration for detection of pink bollworm that were homozygous for resistance to Cry1Ac. On this basis, resistant PBW were detected in 2002 in only 2 out of 13 Arizona strains. The overall frequency of resistant PBW in 2002 for Arizona was 0.17% and ranged from 0.0 to 1.7%. One of six California collections evaluated had a single resistant survivor. No resistant pink bollworms were detected in the single New Mexico and Texas collections evaluated. Resistant PBW were significantly more abundant in Arizona in 2001 and 2002 than they were in 1998, 1999 or 2000. However, the frequency of resistant survivors in bioassays was low for 2001 and 2002, and markedly lower than in 1997. The Arizona Cotton Research and Protection Council evaluated the efficacy of Bt cotton in 2002 using adjacent pairs of Bt and non-Bt fields at 43 locations across Arizona. Pink bollworms were found in an average of 23.3% of these non-Bt boll fields. Bolls from Bt cotton fields yielded an average of 0.144% (range 0 to 1.300%) infested bolls. Of these, all but three of the pink bollworm recovered from Bt cotton plantings came from bolls that tested negative for Cry1Ac. We conclude from these findings that there is no indication that pink bollworm resistance to Cry1Ac was a problem at the locations sampled in 2002. Bt cotton continued to exhibit exceptional field performance in Arizona.

The tolerance of Roundup Ready (RR) Flex cotton to topical Roundup Weathermax (glyphosate) applications and weed management programs in RR Flex cotton were studied in 2002 and 2003 at the University of Arizona Maricopa Agricultural Center. RR Flex cotton demonstrated excellent tolerance to glyphosate as measured by flower pollen shed and lint yield when sprayed topically with glyphosate at 2.25 lb ae/A four times (at the 3 leaf, 6 node, 10 node and 14 node growth stages). Line 1445 containing the current commercial RR genetic construct had a flower sterility/pollen shed rating of 4.6 (1 equals full pollen shed and 5 equals no pollen shed) on 28 June 2003 compared to ratings of 1.1 to 1.9 in lines containing the RR Flex genetic construct. Cotton yields followed a similar pattern with 1145 yielding 386 lb seed cotton/A compared to 1477 to 1894 lb seed cotton/A for the best lines containing the RR Flex genetic construct (yields were generally low because all lines had a Cocker genetic background that is not adapted to hot desert production conditions.) The presence of the RR gene did not affect the yield of genotype pairs that were identical except for the presence or absence of the RR Flex genetic construct. In the weed management study, delaying the first topical glyphosate application resulted in larger, more difficult to control weeds and reduced cotton yield by allowing greater early season competition between weeds and cotton. The best weed control programs included early (1 to 2 leaf growth stage) topical applications at rates greater than 0.75 lb ae/A and a second Roundup application after the first post-planting irrigation (10 node growth stage). The data also suggested that there may be significant value in making a layby, directed-broadcast application that includes a residual herbicide such as prometryn at layby.

A field trial was established during the 2003 growing season as part of the statewide Upland Cotton Variety Testing Program. This trial was located in Thatcher with Dennis Layton Farms as the cooperator. The location was one of eleven around the state. A total of twelve varieties were entered from six cooperating seed companies. Varieties included DP655BR, DP555BR, DP449BR, and DP5690R from Delta and Pine Land Company; FM989BR, FM991R, and FM991BR from Fiber Max; ST5303R and ST5599BR from Stoneville; Riata from CPCSD; AG3601 from Arizona Cotton Growers Association; and SCX-7 from Salcot. The twelve varieties were planted in a randomized complete block design with three replications. Individual plots consisted of 4, 38” rows extending the full length of the irrigation run (1250 ft.). Plant measurements were collected throughout the season to evaluate growth and development characteristics of each variety. Yield and fiber quality data was collected at the end of the season by harvesting and weighing each individual experimental unit. Sub-samples were collected for fiber quality analysis. Percent emergence data indicated differences in seedling vigor and stand establishment. Percent emergence ranged from a high of nearly 90% (Fiber Max FM989BR) to less than 50% (Delta and Pine DP555BR). Even with the low percent emergence for DP555BR an adequate stand was achieved for each variety and did not significantly impact final yield. Significant differences were observed in final lint yields with Fiber Max FM991BR producing the highest yield at 1690 lbs. lint per acre and Fiber Max FM989BR producing the lowest yield at 1292 lbs. lint per acre, a difference of approximately 400 lbs. Delta and Pine DP655BR has been the standard variety planted in the valley for several years. The only variety producing more lint than DP655BR was FM991BR. Lint value calculated using premium/discounts for fiber quality resulted in FM991BR with the highest value at $952/acre. All varieties had overall premiums except for AG3601which was discounted due to high fiber micronaire. Results from this evaluation indicate that FM991BR appears to be an additional variety from which growers have to choose that has the potential to perform very well in the Upper Gila River valley.

This study is a continuation of the variety trials that have been grown in the Duncan/Virden area over many years to supply yield and revenue data on premium cotton varieties for local growers. In recent years, the introduction of herbicide resistant cultivars has been particularly interesting to help clean up many weedy fields. The premium quality New Mexico Acala varieties do not, at this time, have herbicide resistance traits and have struggled to produce high enough yields to compete with the lower quality varieties that are available. Fifteen cotton varieties were tested including two 1517 varieties from New Mexico, Sierra, the newest Roundup-Ready from CPCSD, Salcot Sacala, a new acala from Arizona, and the AZ Cotton Growers variety. The rest of the entries were Roundup-Ready short to mid season varieties from Delta Pine, Stoneville, FiberMax and Paymaster. The highest yielding variety in the trial was Riata, a Roundup Ready Acala from CPCSD, with a yield of 1255 pounds of lint per acre. Sierra, ST 5599RR, ST 5303R and 1517- 99 produced around 100 pounds less lint per acre than Riata but the yields were not statistically different. Plant heights, first fruiting branches (FFB), total nodes and boll weights were measured and height to node ratios were calculated. Many differences were seen between varieties with all of these variables. The values of the variables defining the characteristics of the varieties. HVI data were obtained for fiber qualities of the lint of each variety. This data was then used to determine the value of the lint and then estimate the gross revenue produced by each variety. The highest lint value (cents per pound) was produced by 1517- 99 with 1517-95 and Riata following closely behind. The highest gross revenue was produced by Riata as a combination of the high yield and high lint value.

This study is part of the Regional Pima cotton variety testing program that is carried out in 8 locations from the West Side Field Station in California to El Paso, Texas. Additional varieties are added to the regional standards in this study to supply information needed by cotton growers in southeastern Arizona. Seventeen long staple varieties were tested in a replicated small plot trial on the Safford Agricultural Center in Graham County at an elevation of 2950 feet. Yields were exceptionally high this year with an average yield nearly 600 pound per acre more than was harvested in 2002. The highest yielding variety in this study was Hazera (HA) 195 with a yield of 2027 pounds of lint per acre. This interspecific hybrid , from Israel, has been the highest yielding variety in the study since it was included in 2001. This fuzzy seeded hybrid has fiber qualities as follows: length - 1.41 inches, strength - 34.8 g/tex and uniformity of 87.5. The averages from this study were 1.43, 42.5 and 88.3 for these same variables, respectively. DP HTO was the highest yielding non-hybrid variety in the study with a yield over 1900 pounds of lint per acre. For comparison purposes, it=s length, strength and uniformity were: 1.39, 41.9 and 88.5. Other varieties in the top half of the study were: HA 14-08, OA 360, OA 359, HA 7-66, DP 340 and DP 744, in descending order of their yields. Yield and other agronomic data as well as fiber quality data are contained in this paper along with estimated values of the lint.