Grade reductions for bark contamination constitute costly problem for many U.S. cotton producers. Most of the cotton reduced in grade for bark is grown in the Southwest (Texas and Okla.) where over 85-percent of the cotton is stripper harvested (3). The percent age of cotton produced in this region that is penalized for bark varies greatly from year to year but has averaged about 23-percent over the long term (4). Grade reductions for bark decrease the value of cotton $15 to $20 per bale (4).

Morey et al. (5) have shown that bark contaminants in lint are composed largely of phloem fiber groups separated from pedicles, fruiting branches and main stems of cotton plants. The levels of bark in ginned lint are closely related to the stick content of stripped cotton (8, 9). Environmental factors that influence plant size and conditions at harvest strongly influence stick content and bark contamination of stripper harvested cotton (8, 6, 7). Grade reductions for bark most typically result from stripper harvesting of plants with tall, thick stems and long vegetative branches and which have been subject to field weathering following desiccation by chemicals or freezing conditions.

Recent studies (1, 2) have led to the development of a relatively simple, inexpensive stripper modification that consistently reduced the stick content of stripped cotton and, in some instances, reduced barky grades by as much as 60-percent. The modification requires replacement of the standard, 2-inch wide rubber paddles with narrower (1 to 1.5 inches wide) paddles and timing the stripper rolls to operate brush-to-brush.

Eleven on-farm tests were conducted on the Texas High Plains in 1987 and 1988 to evaluate the operation of conventional and modified farmer owned and operate strippers. An average of 8 and 11 bales were stripped with the conventional and modified harvesters, respectively, at each test site. Harvested cotton was processed at commercial gin facilities and lint samples were submitted to USDA Classing offices for grading. Grab samples were collected from cotton harvested by each machine at the respective test sites and late were fractionated to determine fine trash, hull an stick/stem content. Field loss determinations were made for each harvester at each site. Likewise, the ground speed of the harvesters was determined during each trial.

From the 1986 comparisons it was determined that the modified strippers provided 0.4% higher lint turnouts (20.3 vs. 19.9), 0.6% lower harvest efficiencies (98.0 vs. 98.6), 24% fewer sticks/stems (4.1 vs. 5.4), 8% less fine trash (4.6 vs. 5.0), 14.7% fewer barky bales (25.0 vs. 39.7) and increased net returns to the grower of $1.90 per acre as compared to conventional harvesters. In 1987, the modified strippers provided a 0.1% higher lint turnout (21.5 vs. 21.4), 0.3% lower harvesting efficiencies (98.3 vs. 98.6), 28.5 fewer sticks and stems (2.0 vs. 2.8), 19.7% less fine trash (4.9vs. 6. 1) and increased net returns to the grower of $0.83 per acre. None of the bales harvested in 1987 were downgraded for bark.

In 1986, over 65-percent of the bales classed at the Lubbock Classing office were downgraded for excessive bark content whereas, in 1987, only 15-percent barky bales were reported. These trials confirmed the practicality of using the narrow paddle bat modification to effectively reduce the stick content of stripped cotton and, thereby, the potential for barky grades. Even though the modification resulted in slightly higher field losses, the net returns per acre favored the modified strippers because of the improved lint turnouts.

Field weathering due to delayed harvest can also contribute to bark contamination (7). In 1987, two on-farm harvest-aid chemical trials were conducted to evaluate the effect of timely harvest on yield and fiber quality. The following treatments were applied to non-replicated strips ranging from 10 to 15 acres in size: 1.0 lbs/acre Prep, 1.0 lbs/acre Prep + 0.75 lbs/acre Folex; 0.08 or 0.125 lbs/acre Paraquat and untreated check. Timing of treatments was based on cotton lint model projections (10) and field inspection. Plots were desiccated with 0.6 to 1.0 pt Paraquat after 80 percent (check plots) and 90 to 95 percent (other plots) of the bolls were open. Plots were harvested with farm strippers, the cotton was moduled and later processed at commercial gins. Lint samples were sent to the USDA Classing office at Lubbock for grading. Crop termination with harvestaid chemicals based on cotton lint development model projections reduced yields 5 to 13%, mainly because of unusually favorable fall growing conditions. In one test, the Prep and Prep + Folex reduced barky grades 27%. At the other test site, the same treatments actually increased the incidence of barky grades because the treated plots were harvested before the cotton was properly conditioned following desiccation.

Under normal fall conditions the use of harvest-aid chemicals and proper harvesting procedures can minimize weathering losses and the potential for grade reductions due to bark.