In further work on the byssinosis problem, fiber of seven varieties of cotton was field-weathered in 1985 at two sites near College Station, TX, one an upland site and the other a bottomland site. Well-fluffed locks were collected representing three weathering periods, and counts were made for total, gram-negative, and presumptive coliform bacteria on the fiber. All counts on fiber from the first collection (Sept. 3-4) at both sites were in a low range, with counts on the fiber from the bottomland being consistently higher than those on fiber from the upland site, and with counts on fiber from the lower part of the plants at the bottomland site exhibiting higher counts than fiber from the top. There were no clear varietal differences. At the second collection (Oct. 3), the general level of counts had increased greatly at both sites, with fiber from the upland site exhibiting consistently higher counts than fiber from the bottomland; no consistent differences in counts between fiber from the bottom and top of the plant were measured. There were no varietal differences. At the third collection (Nov. 6), the general level of counts at the two sites remained at about the same level as in the second collection, the fiber from the upland site again exhibiting counts consistently higher than counts on fiber from the bottomland, but again with no consistent differences in counts between fiber from the bottom and top of the plants. Again there were no varietal differences. The overall average fiber pH's were 6.6, 8.6, and 7.8 for the first, second, and third collections, respectively. Implications of the data in respect to the byssinosis problem are discussed.