Reduction of populations of the B strain (poinsettia) of sweetpotato whitefly (SPWF), Bemisia tabaci Gennadius, resulted from applications of buprofezin or amitraz to cotton in central Arizona (Maricopa, AZ). Control was fair to good control for this insect. However, yields of seed cotton in treated plots were not increased significantly compared to untreated plots, following three applications of these insecticides during the season. Similar results on percentage sugar on lint and lint stickiness from honeydew of SPWF in both treated and untreated plots were obtained.

Four field experiments were conducted to evaluate efficacy of a number of registered or experimental insecticides against the sweet potato white fly. Several materials gave encouraging results under unusually high population densities. Among the more effective treatments in one or more tests were: Capture, Danitol, NTN 33893, pyriproxyfen and SN 85292, and combinations of Capture +endosulfan, Capture +Ovasyn and Danitol +Orthene.

The sweet potato whitefly (SPWF) Bemisia tabaci (Gennadius). has become a serious problem of a number of agricultural crops in the southern tier of states in the U.S. In the southwestern U.S. it seriously affects summer crops such as melons and cotton, and fall, winter and spring vegetable crops such as lettuce, broccoli and cauliflower. Since this insect has no overwintering resting stage a succession of host plants is necessary in order to span the gap from cotton season to cotton season. This study characterized seasonal population trends of SPWF in cotton and then identified subsequent hosts which were important in the overwintering survival of this insect. The "off-season" hosts included certain weeds as well as cultivated crops and all appear to be important in the seasonal population dynamics of this whitefly.

Four Pima cotton entries were grown in large plots at MAC, not treated with insecticide, to determine their response to attack of pink bollworm. P62, an early -maturing germplasm line, sustained significantly less seed damage caused by pink bollworm than did 'Pima S-6', Pima S-6 nectariless, and P62 nectariless. The nectariless trait did not reduce seed damage significantly.

In 1991 four Insect Pest Management studies were conducted to help understand the pink bollworm and aid in its control. The first two were replicated cage studies where cages were placed over the soil and insect emergence from the soil was monitored several times a week from January to mid-summer. The first had pink bollworm (pbw) infested bolls buried at 0, 2, 4 and 8 inches under the soil. The second had four plow down dates and sub -treatments of one or no irrigations. Perhaps due to the cold winter, few pink bollworm moths emerged in either study. In the buried infested boll study, no bolls buried at 8 inches produced pink bollworm moths in the spring and few emerged from either 4 or 2 inches. Of the bolls left on the surface, there was emergence from only some of the replications. Less than 1% of all of the potential moths emerged. In the plowing test, only 6 pink bollworms emerged in any of the 32 cages placed in the plots over the 18 weeks of the study. It appears to be a numbers game and the chance of having a pink bollworm emerge under a randomly placed cage in a field is very small. The second two tests involved Trichogramma bactrae, a trichogrammatid species imported from Australia. The first of these two studies involved placing laboratory produced pink bollworm eggs in a pattern around a release site for the parasitic wasps to determine their area of influence. This was done on three different occasions with varying success. The study was complicated by the fact that the wasp hatch must be coordinated with the pink bollworm egg deposition. Never-the-less, some pink bollworm egg parasitization took place. This study needs to be repeated. The second of these studies was two plots side-by-side, one treated with chemicals to control pink bollworm (and other insects), the other with weekly trichogramma releases during August and September. At the end of the season, 20 plants were removed from each plot and boll infestation and boll load were determined. The parasite controlled plots had a higher infestation level (35% to 23% with an LSD(05) of 9.9) than the chemically treated plots. There was no difference in the boll load between the two treatments. This study should be continued, preferably with a check plot so the value of the trichogramma can be evaluated more accurately.

The computer program AZSched, with weather data obtained from AzMet, was used to schedule irrigation for a yield trial of Upland Cotton (DPL 90) at the Maricopa Agricultural Center in 1991. Cotton lint yields were compared between plots from four treatments involving the combination of two irrigation efficiencies (70% and 90 %) and two nitrogen fertilizer applications (broadcast and sidedress). The amount of nitrogen fertilizer applied to the broadcast plots was 145# N /A. One hundred thirty pounds NIA was applied to the sidedressed plots. A potassium bromide tracer was applied to select areas in each plot at the time of initial fertilization. Soil samples from each plot were taken to a depth of 10' for analysis of bromide and nitrate to determine the depth of water movement through the soil profile. Irrigation amounts averaged 42.8" for 70% efficiency and 36.7" for 90% efficiency. No significant difference in lint yield was noticed between the two nitrogen fertilizer applications, but significant differences existed between the two irrigation efficiencies.

Irrigation scheduling software has been developed that is menu driven, user friendly and capable of scheduling up to 60 fields. This software is demonstrated scheduling irrigation on both Pima and upland cotton in this paper.

Pima experimental strains P73, P74, P75, P76, P77, and P78 and the varieties Pima S-6 (PS-6) and Pima S-7 (PS-7) were grown in Regional tests at seven locations across the Pima belt in 1991. When yields were averaged across five test locations below 2,500 feet, the strains P73, P74, P75, and P76 ranked higher than PS-6, but lower than PS-7. At two test locations above 2,500 feet the strains P76 and P73 ranked higher in yield than PS-6 and PS-7. There was a trend across all locations for the strains P73, P74, P75 and P76 to be shorter in height than PS-6. The strain P76, which was the highest yielding strain across all locations, was the shortest strain at five of the seven locations. Strains included in all tests tended to have whiter, stronger fiber than PS-6. The above was not the case for P77 and P78, which were present in only two tests.

Three new California acala varieties are evaluated in this field trial along with nine New Mexico acalas. The highest yielding variety was New Mexico's 1517-88 with a yield of 2.8 bales per acre. California's MAXXA came in number 3, but not very far behind, yieldwise. Nazas 77, a rust resistant variety from Mexico,was evaluated in a second trial at this same location. Rust was not a problem at this site in 1991 so its rust- resistance could not be evaluated, but it yielded within 5% of the yield of 1517-88.

Weather played a major part in the production of cotton in the Safford valley in 1991. A cool spring and early summer slowed the development of cotton and gave an advantage to varieties that could produce quickly in the warm late summer months. Three Stoneville varieties came to the top of the test with Stoneville 506 producing the highest yield of over 4000 pounds of seed cotton per acre. The New Mexico acalas, 1517-91 and 1517-88 also produced well, and depending on the premium, could produce more income per acre than the top varieties. Fifteen varieties were tested at this site and all produced over 2 bales per acre.