Introduction

Cotton fiber quality is no longer an afterthought; it is becoming an increasingly important issue for Georgia producers. The previous mindset concerning cotton production has been to produce a crop so that yields were maximized and minimum quality standards were met. Research and breeding efforts in the past have also focused primarily on yield enhancement. However, current changes in spinning technology and the global nature of the cotton market have necessitated the production of higher quality fiber. Basic knowledge of fiber development and how management practices enhance or preserve fiber quality are now a necessity. In the near future it is possible that the production of a cotton crop will no longer be deemed a success based solely on the number of bales produced per acre. A full understanding of the fiber development process, as well as an appreciation of the agronomic and environmental factors affecting fiber quality is imperative in order for this issue to be intelligently addressed. The purpose of this bulletin is to describe fiber development on a physical level and to delineate what we do and do not know about the agronomic and environmental factors affecting fiber quality.

The Georgia Situation

Prior to the late 1990s and the early 21 st century, the fiber quality of cotton produced in Georgia was second only to the quality produced in California, where genetically superior Acala varieties are grown in a Mediterranean-like environment. Recently, certain fiber quality parameters of Georgia-grown cotton have not only fallen well below Western U.S. cotton, but also below the average fiber quality produced in other southern and southeastern states. The fact is that certain fiber quality parameters may fall short in a given year for any location, and this cannot be prevented. The current concerns regarding fiber quality in Georgia involve more than year to year aberrations. An examination of the quality factors on a purely numerical basis indicates that several trends exist, but they do not tell the whole story nor do they explain why there is now a selected U.S. mill bias against Georgia cotton.

During the first half of 2004, several mills across the U.S stated intentions to limit or exclude the purchase of cotton produced in the state of Georgia. Many discussions have been held trying to isolate what fiber quality parameter is the culprit for these biases, and the general consensus is that problems are associated with increased levels of short fibers (those less than 0.5 inches in length) in bales produced in Georgia.

Basic Physiology of Fiber Development in the Cotton Plant

Cotton fiber development begins at anthesis, or pollination of the flower (Benedict, 1984). Generally, a cotton flower is pollinated the day it opens. The ensuing development of the cotton fiber is a very ordered process, consisting of four specific stages: initiation, elongation, secondary wall synthesis, and maturation.

Initiation

Fiber initiation occurs as the epidermal cell expands above the surface of the ovule, or developing seed. While initiation first occurs at pollination, this process may occur in several waves, making it possible to find fiber initiatives for 5 to 10 days after pollination has occurred (Fig. 1) (Stewart, 1975; Lang, 1938).

Figure 1. Fibers elongating from the seed-coat of a developing seed on the day of anthesis. From these pictures it is apparent that not all fibers are initiating at the same time or at the same rate. Thus this process may occur over several days. During this stage, fibers are hollow and growing as water and carbohydrates are being delivered inside them. (Pictures courtesy of Mac Stewart, University of Arkansas)

Anatomically speaking, these fibers are seed hairs or single hyperelongated cells arising from the outer layer of the seed coat. During this stage fibers are hollow, almost resembling a straw or tube with a capped end (Stewart, 1975).

Fiber Elongation

Fiber elongation is the stage during which the fiber is lengthening, and occurs as a thin cell wall of carbohydrate polymers is deposited allowing the fiber to elongate (DeLanghe, 1986). Water pressure inside of the developing fiber in conjunction with the deposition of carbohydrate polymers drives elongation (Fig. 2). The elongation period of fiber development depends on the species of cotton. The elongation phase of long staple Pima ( Gossypium barbadense ) varieties of cotton may last for 30 to 40 days. In Upland ( Gossypium hirsutum ), or short staple cotton varieties, such as those grown in Georgia the elongation period lasts for approximately 3 weeks (Schubert et al., 1976). At the end of the elongation phase the fiber still resembles a hollow tube or straw.

Figure 2. Fibers elongating A) 3 days and B) 4 days after flowering. (Pictures courtesy of Mac Stewart, University of Arkansas)

Fiber length, reported as staple or the upper one-half mean length (and to some extent fiber strength and uniformity), is influenced during this stage of fiber development. While fiber length is definitely under significant genetic control, other agronomic and environmental parameters can also influence fiber length.

Secondary Wall Synthesis

During secondary wall synthesis or thickening, cellulose is deposited inside the elongated fiber. This process contrasts the way a tree trunk grows in thickness, wherein new growth, or carbohydrate, is deposited along the outside of old growth. Thickening of the fiber begins once the boll is approximately 15 days old and occurs for approximately 3 weeks. There is some overlap in transition between elongation and fiber thickening; in other words, the fiber is both elongating and thickening simultaneously during the period of 2 to 3 weeks post-flowering (Fig. 3) (Schubert, 1975 and Schuber et al., 1976).

The degree of secondary wall thickening is measured via the quality parameter micronaire. This parameter is under some genetic control; but it is heavily influenced by environment and management practices.

Figure 3. Fiber length and thickness as a function of boll age of upland ( Gossypium hirsutum ) cotton.   Fibers elongate for approximately the first 21 days after anthesis, thickening begins at about 15 days after anthesis and continues for approximately 3 weeks. ( Graph adapted from Shubert, 1975 and 1976 )

Maturation

The previous three phases of fiber development occur while the fiber is alive and actively growing. Maturation of the fiber occurs after the boll has opened and is best described as the drying of mature metabolically inactive fibers.

While there is no quality measurement directly related to the maturation process, it is important to understand that the fiber quality within a boll is at its utmost on the day of boll opening. Data will be presented to indicate that harvesting bolls as close to physiological maturity as possible enhances the quality of the crop produced.

What is Fiber Quality?

Fiber quality is a general term given to a set of measurements that describe a sample of fibers extracted from a bale of cotton (Bradow and Davidonis, 2000). These measurements are then compared to a set of United States Department of Agriculture (USDA) standards and are used to determine price premiums and discounts for the bale from which the sample was taken. The measurements taken include: length, uniformity, strength, micronaire, color grade, trash, leaf grade, preparation, and extraneous matter (USDA, 2001). In the following sections of this bulletin, each of these parameters will be discussed as to how they are measured, what factors influence them, how they may be manipulated, and what concern they are to the textile industry.

All fiber quality parameters listed previously, with the exception of leaf grade, extraneous matter and preparation, are determined via High Volume Instrument (HVI) analysis. Practically all cotton grown in the U.S. is classed by the USDA through HVI testing at the request of producers. While not mandatory, classification is generally essential in marketing the crop to buyers and participating in the USDA price support system. HVI data are obtained from two 4 oz samples of fiber taken from each side of a ginned bale of cotton (Fig. 4).

Figure 4.   Fiber samples to be classed at the USDA classing office in Macon, GA.   These are 4 oz samples taken from two sides of a ginned bale of cotton.   HVI analysis will be performed on these samples at the classing office.

During development cotton fibers are livings cells. Like all living cells, these fibers develop and respond individually to fluctuations in the environment as a whole and fluctuations in the physiology and morphology of the cotton plant. For instance, fibers on a single seed may vary in length, shape, thickness, and maturity. In addition, variations within the plant, among parts of a field, and between plants exist. Thus, each bale of cotton contains a highly variable array of fibers (Bradow and Davidonis, 2000).

Fiber Length/Staple

Fiber length is determined from a "beard" of cotton fibers. The beard is formed from a sample of cotton fibers that are grasped by a clamp then combed and brushed to straighten and make the fibers parallel (Fig. 5). The length reported is the average length of the longer one-half of the fibers, also called the upper half mean length or staple (Fig. 6). Staple is reported in 32nds of an inch (USDA, 2001). See Table 1 for conversions from inches to staple. Fiber length and staple can be influenced by several factors including variety, temperature, water stress, nutrient deficiencies, and ginning practices.

Figure 5. Fibers from each bale sample are "combed" and straightened into a beard. This beard is then used to make length, strength and uniformity measurements. Note the unevenness of the sample illustrating the inherit variability of individual fiber lengths in the sample.

Figure 6. Staple, which is   reported on the classing document, is the upper 1/2 mean length of the fibers in a sample. Thus, in this example of 10 fibers, staple would be the average length of the longest 5 fibers. The mean length of all 10 fibers is important for determining the uniformity ratio, which is calculated by dividing the mean length by the upper ½ mean length and multiplying by 100. Uniformity Index = (mean length/upper ½ mean length) x 100 The Uniformity Index will always be less than 100, due to the inherent variability of fiber lengths.

Both individual fiber length and staple are important to the textile mills for several reasons. Longer fibers produce stronger yarns and fabric and enhance fabric appearance. These longer fibers are also necessary for the textile mills to maintain and enhance yarn processing efficiencies (USDA et al., 2002). Air jet spinning is currently the most productive and efficient spinning technology utilized by textile mills and is the most sensitive to short fibers. While air jet spinning requires longer fibers than older technologies such as ring spinning, this type of spinning technology is utilized to produce products such as low-end bed sheets and fleece in sweat shirts (USDA et al., 2002).

Impact of Varieties on Fiber Length and Staple

The most dramatic difference noted among varieties concerning fiber length occurs when Pima and Upland cottons are compared. The staple of Pima cotton typically ranges from 36 to greater than 42, while the staple of upland cotton produced in Georgia ranges from 32 to 36 (USDA, 2001). The differences among Upland cotton varieties are typically not as dramatic as those between differing species of cotton, yet they are no less important. Figures 7 and 8 graphically depict the range in staple between cotton cultivars evaluated in the 2003 University of Georgia statewide variety trials.

Figure 7. Average staple of 2003 dryland Univeristy of Georgia Statewide Cotton Variety Trials. A)Early maturing varieties; B)Late maturing varieties.

Figure 8. Average staple of 2003 irrigated Univeristy of Georgia Statewide Cotton Variety Trials. A)Early maturing varieties; B)Late maturing varieties.

Temperature and Fiber Length

The base temperature for cotton growth is 60 degrees F (Mauney, 1986). Studies in the High Plains of Texas have indicated that nighttime temperatures near this base temperature delayed fiber elongation by 4 to 5 days compared to nighttime temperatures of 77 degrees F (Gipson and Ray, 1968, 1969). Heat units or growing degree days also influence fiber elongation, and this relationship may be further modified by the relative maturity of the cultivar grown. It has been demonstrated that increasing season-long heat units increased the short fiber content of later maturity cotton cultivars and decreased the short fiber content of earlier maturity cultivars (Bauer and Bradow, 1996; Bradow and Bauer, 1997). The majority of cotton produced in Georgia is of later maturity varieties and thus it may be argued they are more sensitive to high ambient nighttime temperatures often encountered in Georgia.

Water Stress Effects Fiber Length and Staple

As with many other agronomic factors, water relationships and irrigation practices have been studied primarily in relation to yield. One study conducted in the early 1980s indicated that fiber length was not impacted until water stress was such that yields were limited to less than 630 lb/A (Grimes and Yamada, 1982). Other studies have demonstrated that water stress early in the bloom period was less detrimental to fiber length than water stress late in the bloom period (Marani and Amirav, 1971; Shimishi and Masani, 1971; Hearn, 1976). Sensitivity of fiber elongation to severe water stress is apparently due to the physiological and mechanical processes of cell expansion (Hearn, 1994). This relationship is clouded by the fact that cotton blooms over an extended period of time. Thus some fibers may be elongating during a drought period while others elongate during a period when moisture is more plentiful (Bradow and Davidonis, 2000).

During the years of 2001 and 2002, fiber quality data for Georgia were evaluated by large regions across the state (Jost unpublished, 2002). This process was utilized to maintain anonymity of producers and gins. The season long average staple for these regions was then compared to season-long rainfall. There was a significant relationship between average staple produced in a region and rainfall received in that region. Areas which received higher rainfall produced longer fibers (Fig. 9). This supports the idea or understanding that irrigation may play a role in enhancing fiber length.

Influence of Rainfall on Average Staple

Figure 9. Staple measurements were averaged for 4 regions within the state of Georgia in 2001 and 2002. The average staple was then correlated with rainfall received from May through September in those regions (Jost, 2003. unpublished data).

Impact of Fertility on Fiber Length

It has long been recognized that fertility levels significantly affect lint yield. However, the impact of fertility on fiber quality, especially staple length, is much less concrete. Research results are often contradictory due to the confounding effects of genotype, climate, and soil conditions. Laboratory studies have shown that malate and potassium (K) levels increase during rapid fiber elongation and fall as the rate of elongation decreases (Basra and Malik, 1983). In addition, recent work in Arkansas has shown that K is especially important during fiber initiation and the early elongation phase of fiber development (Fig. 10). Studies examining varying levels of K fertilization do not present any conclusive results (Bradow and Davidonis, 2000). Present concerns about the length and uniformity of Georgia cotton suggest the need for reevaluation of the relationship of fertility to fiber quality.

Figure 10. K has been shown to accumulate in fiber initiatives as they begin to elongate. (Pictures courtesy of Mac Stewart, University of Arkansas )

Defoliation and Harvest Timing Effects on Fiber Length

As stated previously, fiber length is determined during the first three weeks after pollination of the flower. Therefore, when the crop is mature enough to defoliate, the fiber lengths of the vast majority of the bolls on the cotton plant should already be established. However, studies conducted at the University of Georgia indicate that defoliation and harvest timing do have an effect of fiber length of the harvested crop. These results demonstrated that the mean fiber length as measured by the Advanced Fiber Information System (AFIS) was greatest when harvest aids were applied between 39.1 and 56.7% open boll (Figure 11). Delaying defoliation beyond this point reduced fiber length (Bednarz et al., 2002), and one must assume that the reduction in fiber length was due to fiber degradation as a result of weathering and ginning rather than from some physiological process.

Figure 11. Effect of delayed defoliation and harvest timing on average fiber length as measured via the Advanced Fiber Information System (AFIS). Defoliants were applied weekly which corresponded to various % open boll measurements. Plots were harvested 2 weeks after defoliant application (Bednarz et al., 2002)>

Economics of Fiber Length

The base fiber length according to the current USDA Cotton Loan Schedule is 34 (USDA, 2004). Staple lengths below 34 incur discount while those above 34 are eligible for premiums. The level of premium or discount is also related to color. The higher the color grade the lower the discount. For example a 33 staple with a base color grade of 41 and leaf of 4 would incur a discount of $0.0205/lb. A 33 staple with a lower color grade of 51 and leaf of 4 would incur a discount of $0.0385/lb.

How Does Georgia Compare

In three of the past six years, the average staple of the Georgia crop has been the shortest compared to other classing offices in the southeast (Florence, SC; Memphis, TN; and Birmingham, AL). In only one of those years did Georgia cotton have the highest staple among the classing offices in the southeast (Fig. 12). A more startling number to examine is the percentage of the Georgia crop classified as short staple (less than 34). During the years from 1993 through 1997 less than 15 percent of the Georgia crop was classed as short staple. From 1998 to the present in only 1 year was less than 20 percent   of the crop classed as short staple, and in two of those years, more than 30 percent of the cotton produced in Georgia was classed as short staple (Fig. 13).

Figure 12. Average staple by year as reported by Southeastern USDA cotton classing offices.

These data have spurred some interesting discussion and theories. Prior to the late 1990s, the predominate variety planted in Georgia was DPL 90 which was noted for superior staple and strength. During the late 1990s a major shift was made away from DPL 90 to transgenic varieties, those containing the Roundup Ready ® and/or the Bt genes. Therefore, the argument has been made that this shift in varieties is the sole cause of this reduction in staple observed in Georgia. This shift in varieties also coincided with the onset of a lingering drought, which has also been shown to affect staple. The 2003 production season was uncharacteristically wet and cool, yet staple problems persisted. Thus the current situation is probably in part a function of both environmental fluctuations and variety selection.

Figure 13. Short (less than 34) staple bales as a percentage of total bales produced in Georgia from 1993 though 2004.

Uniformity

Uniformity is related to fiber length in that the uniformity index is the ratio between the mean length and the upper one-half mean length of the fiber sample (refer to Fig. 6) (USDA, 2001). This measurement is reported via HVI analysis as a percentage. Thus, if all the fibers in a sample are the same length the uniformity index is 100. A large range of fiber lengths in the sample will result in a lower the uniformity index. Due to the natural variation of fiber lengths in a cotton boll, the uniformity index is always less than 100 (USDA, 2001). The Uniformity Index generally ranges from 77 to 85.

Uniformity is important to the textile industry because it is an indirect indicator of short fiber content, which ultimately affects yarn evenness, strength, and the efficiency of the spinning process (USDA, 2001). It is important, again, to note the difference between short staple and short fibers. Short staple cotton is that which has an average staple (upper ½ mean length) of less than 34. Short fiber is a term given to individual fibers that are less than 0.5 inches in length. Since the uniformity index is determined via two length measurements, those parameters that influence fiber length can also influence the uniformity index.

Variations in fiber length occur on a single seed. Studies have demonstrated that fiber lengths were shortest at the micropylar (pointed) end of the seed (Vincke, 1985), however, this is also where the most mature fibers with the largest perimeters were found (Fig. 14). The variation of fiber length on an individual seed is further modified by the location of the seed in the boll. Typically, seeds located near either the basal or apical ends of the boll produce the shortest fibers, while those near the middle produce the longest fibers (Porter, 1936).

The location of the boll within the canopy has also been shown to influence fiber length (Bradow and Davidonis, 2000). Fibers from bolls produced near the middle of the canopy (fruiting branches 9, 10, and 11) are consistently longer than fibers produced lower in the canopy (fruiting branch 7).

Collectively, these observations indicate that differing microenvironments have a profound impact on fiber length despite the fact that all fibers are produced in the same production season, in the same field, and on the same plant.

These data have spurred some interesting discussion and theories. Prior to the late 1990s, the predominate variety planted in Georgia was DPL 90 which was noted for superior staple and strength. During the late 1990s a major shift was made away from DPL 90 to transgenic varieties, those containing the Roundup Ready ® and/or the Bt genes. Therefore, the argument has been made that this shift in varieties is the sole cause of this reduction in staple observed in Georgia. This shift in varieties also coincided with the onset of a lingering drought, which has also been shown to affect staple. The 2003 production season was uncharacteristically wet and cool, yet staple problems persisted. Thus the current situation is probably in part a function of both environmental fluctuations and variety selection.

Figure 14. Cross section of a boll at maturity. The lengths of fibers produced within a boll vary by seed location. The lengths of fibers also vary by location on the seed.

Many factors influence fiber length, and some of these undoubtedly affect length uniformity. Recent indications are that variety selection may also play a role. Fiber quality from the University of Georgia statewide variety trials is assessed via ginning hand-picked samples on a table-top gin and therefore does not allow for a "real world" look at many fiber parameters, including uniformity. Thus, our knowledge base for this factor is extremely limited. The effects of temperature, water stress, and soil fertility on fiber uniformity are also basically unknown at this point.

Defoliation and harvest timing have an effect on short fiber content, as indicated via AFIS analysis (Bednarz et al., 2002). As harvest timing is delayed, short fiber content increases (Fig. 15). The more fibers present less than 0.5 inches in length in a sample, the lower the uniformity index. Again, this is most likely due to environmental factors and fiber deterioration rather than a physiological factor.

Figure 15. The effect of delayed defoliation and harvest timing on short (<0.5 in) fiber content. Defoliants were applied weekly which corresponded to various percentages open boll measurements. Plots were harvested 2 weeks after defoliant application. As defoliation and harvest timing were delayed short fiber content increased. This measurement can be correlated to length uniformity which is an indirect measure of short fiber content. As short fiber content increases length uniformity decreases (Bednarz, et al., 2002)

The base range for uniformity is 80 to 82, with uniformity indices greater than 82 eligible for premiums and those lower than 80 receiving discounts (USDA, 2004). On a purely monetary basis the discounts for uniformity are not as severe as those for staple. For example, a uniformity of 79 receives a discount of only $0.0055/lb. However, as mentioned previously, uniformity is an indirect measurement of short fiber content (fibers less than 0.5 inches). This appears to be one of the parameters for which mills are increasing scrutiny and may be a source of some of the bias against Georgia cotton.

A comparison of other southeastern classing offices shows some remarkable trends for uniformity (Fig. 16). During the past five years Georgia cotton has consistently been among the lowest in terms of uniformity. While the cause of this discrepancy in uniformity between Georgia and other southeastern states is not certain, one line of thinking relates to differences in cropping systems between these locations. Georgia has a much greater ratio of peanut to cotton acres. The common practice of producers is to dig and harvest peanuts prior to cotton harvest, because peanut yields decline significantly when digging and harvesting are delayed beyond maturity. The peanut-cotton issue has become more significant as peanut planting has been moved later into May due to concerns over Tomato Spotted Wilt Virus. Now cotton and peanuts are planted at approximately the same time, the end result being that cotton is ready to defoliate and peanuts are ready to dig simultaneously.

Figure 16. Average uniformity by year as reported by Southeastern USDA cotton classing offices.

Relationships between Staple and Uniformity

A recent survey was conducted of several gins across Georgia. While several trends were noted in this survey, it must be understood that a "gin" is more than a representation of ginning practices. A "gin" is representative of all management factors of producers utilizing that gin, the environment in that locale, and ginning practices together (Shurley, 2004)

The average staple by gin in 2003 is depicted in Figure 17; this figure readily illustrates the variability in staple across the state, ranging from 33.5 to nearly 35. More alarming is the percentage of bales by gin classed as short staple, or less than a 34 staple (Fig. 18). While some gins had a very low percentage of short staple bales, others surpassed 40 to 50 percent. This short staple problem may be related to low uniformity issues. Figure 19 illustrates a positive relationship between percent short staple and low uniformity bales.

Figure 17. Average staple produced by gins across Georgia in 2003. A "gin" is representative of all management factors of producers utilizing that gin, the environment in that locale, and ginning practices together (Shurley, 2004).

Figure 18. Percent of bales classed as short (<34) staple produced by gins across Georgia in 2003. A "gin" is representative of all management factors of producers utilizing that gin, the environment in that locale, and ginning practices together (Shurley, 2004).

Figure 19. Relationship between short (<34) staple and low (<80) uniformity bales produced by gins across Georgia in 2003. A "gin" is representative of all management factors of producers utilizing that gin, the environment in that locale, and ginning practices together (Shurley, 2004).

Strength

Fiber strength measurements are reported in grams per tex. A tex unit is equal to the weight in grams of 1,000 meters of fiber. The strength that is reported is the force in grams required to break a bundle of fibers one tex unit in size (USDA, 2001). The fiber strength measurement is made by clamping, pulling, and breaking a bundle of fibers, with a 1/8-inch spacing between the clamp jaws (Figure 20).

The breaking strength of the individual cotton fibers is considered to be the most important factor in determining the strength of the yarn spun from those fibers (Munro, 1987; Patil and Singh, 1995; Moore, 1996). Again, shifts in spinning technology have placed a premium on individual fiber strength, as these new technologies produce yarns of lower strength (Patil and Singh, 1995).

Figure 20. The "beard" of fibers above is utilized to measure length, uniformity, and strength. The beard on the left has been broken to obtain the strength measurement.

Conflicting results have been observed concerning environmental influences on fiber strength. Studies as early as 1956 showed positive relationships between temperature and canopy sunlight absorption and fiber strength. It was also shown that fiber strength increased with decreasing rainfall (Hanson, 1956). In general, conclusions from this study indicate that fiber strength may possibly be more responsive to the environment than length and fineness (discussed later). Current studies have shown that genotype or variety may be a more determinative factor of strength than environment (Mackenzie and Van Schaik, 1963; Green and Human, 1983; Green and Culp, 1990; Smith and Coyle, 1997).

Recent studies have indicated that there is a negative relationship between lint yield and fiber strength. Studies completed in 1997 have shown that both fiber strength and length were negatively correlated with some basic yield components. More specifically, these studies demonstrated that fiber weight and numbers decreased as fiber strength increased (Coyle and Smith, 1997; Smith and Coyle, 1997).

Correlations with fiber strength and heat unit accumulation during the flowering period have also been established. One study indicated that fiber strength was greatest from bolls that developed from flowers produced during the first 4 to 6 weeks of flowering, while flowers that opened during the latter two weeks of the flowering period produced bolls with the lowest fiber strength (Jones and Wells, 1997). Increasing heat unit accumulation was surmised to cause the increase in fiber strength.

Due to the physiology of boll development, one might expect that early defoliation and premature application of boll openers would reduce fiber strength. Once defoliants are applied, the function of the leaf essentially ceases; thus, cellulose deposition into the elongated fiber cell is inhibited, limiting fiber strength. However, studies have shown that as early as 20 days after pollination, a fiber bundle reaches 50 percent of the mature fiber bundle strength. By 30 days after pollination more than 80 percent of the mature fiber strength had been attained. In addition, at 30 days after pollination only 60 percent of the mature micronaire, or fiber thickness, value had been attained. (Micronaire is a measure of fiber thickness and maturity and will be discussed later.)   In summary, fiber strength is developed well in advance of final fiber maturity and ultimate fiber thickening (Lewis and Benedict, 1994).

Field studies examining early defoliation timings have confirmed these laboratory results. Snipes and Baskin (1994) demonstrated that defoliation at 20 percent open boll actually increased fiber strength and length as compared to later defoliation timings. Bednarz et al., (2002) also demonstrated these results in Georgia where fiber strength was maximized at defoliation timings as early as 6 percent open boll. Of course, the impact of early defoliation on fiber maturity is a much more complex issue and must also be considered. Yield is also compromised with ultra-early defoliant applications (Snipes and Baskins, 1994; Bednarz et al., 2002)

The base range for fiber strength is 25.5 to 29.5. Discounts for strength can range from $0.0011/b to $0.005/lb.

Generally fiber strength is not an issue for Georgia producers. Typically, less than 10 percent of the Georgia crop receives deductions due to strength. In years when significant weathering occurs due to late season rainfall, this percentage may increase. However, concerns about reduced fiber strength may be larger than what can be explained via monetary deductions, especially if fiber strength is reduced enough to allow ginning to compromise uniformity and increase short fiber content.

Micronaire

Micronaire is a measure of fiber fineness and maturity (USDA, 2001). Fiber maturity means different things to cotton processors, physiologists, and producers. To a physiologist and a cotton producer, cotton maturity indicates a measure of time elapsed between flowering and harvest. To a textile manufacturer, maturity indicates fiber wall thickness (Bradow and Davidonis, 2000). Immature fibers contain thinner walls than more mature fibers of the same variety. While a finer mature fiber results in more fibers per unit area and thus a higher luster dyed fabric (Ramey, 1982), a fine but immature fiber will stretch and tangle, causing problems in the spinning process. In addition, dye uptake will also be inconsistent with an immature fiber (Bradow and Davidonis, 2000). No matter the process influenced, the important factor is fiber wall thickness.

The thickness of the fiber, and consequently the fiber wall, is measured via resistance to airflow (USDA, 2001). Micronaire is an indirect measure of the air-permeability of a fiber sample of known mass enclosed in a container of fixed dimensions (Fig. 21) (Bradow and Davidonis, 2000). Air-permeability is negatively correlated with surface area. Thus the more surface area in a given sample the less air flow and the lower the micronaire. Less surface area allows for more air flow and elicits a higher micronaire reading.

The relationship between micronaire and air flow in the enclosed chamber can be better explained via a discussion of fiber thickness. The thicker a fiber is the more it weighs, thus fewer of these fibers are required to constitute the 10 g sample necessary for the air flow test. The thinner a fiber is the less it weighs, thus more of these fibers are required to constitute the 10 g sample. Air flow decreases as fiber number and subsequent surface area increases. Therefore, high micronaire is associated with thick, more mature fibers. Low micronaire is associated with thinner, less mature fibers.

Figure 21. The micronaire chamber. Approximately 10g of fibers are placed into this chamber. The resistance of this sample of fibers to airflow is measured and is reported as micronaire.

Previous research has indicated that genotype or cultivar had a limited impact on micronaire, and concluded that environment is much more of a determining factor (Bradow and Davidonis, 2000). However, examinations of recent University of Georgia Statewide Cotton Variety Testing results reveal that there are significant differences in micronaire among commercially available varieties (Figs. 22 and 23). If breeding programs are focused solely on yield with little consideration for fiber quality, higher micronaire cottons could be selected, since thicker fibers weigh more.

Figure 22. Average micronaire of 2003 dryland Univeristy of Georgia Statewide Cotton Variety Trials. A)Early maturing varieties; B)Late maturing varieties.

Figure 23. Average micronaire of 2003 irrigated Univeristy of Georgia Statewide Cotton Variety Trials. A)Early maturing varieties; B)Late maturing varieties.

As stated previously micronaire is a measure of fiber wall thickness. A fiber thickens as cellulose is deposited inside the elongated fiber cell. Cellulose is derived from photosynthetic carbon fixation. Thus any factor that influences carbon fixation ultimately influences micronaire.

Temperature -- more specifically, heat unit accumulation -- has a profound effect on micronaire. Studies conducted in South Carolina, South Africa, and Louisiana, all demonstrated a significant relationship between increased heat unit accumulation and increased micronaire (Aguillard et al., 1980; Greef and Human, 1983; Porter et al., 1996). These studies generally manipulated heat unit accumulation by staggering planting dates so that later planting accumulated fewer heat units and produced lower micronaire readings. However, a similar observation was made with an early planting where suboptimal temperatures were encountered during the early season (Bradow et al., 1997).

The effects of available water on fiber maturity and micronaire are not as clear as other factors. An over-supply of water can delay cellulose deposition by stimulating excessive vegetative growth and lessening the carbohydrate supply for developing bolls (Hearn, 1994). However, adequate water during mid-season can increase cellulose deposition and micronaire by enhancing carbon fixation. Studies have also indicated that delaying irrigation initiation may increase micronaire, while timely irrigation and water conservation techniques may decrease micronaire (Spooner et al., 1958; Singh and Bahn, 1993). Rainfall may decrease micronaire in several ways. An adequate water supply allows increased boll set at upper and outer fruiting positions where less mature fibers are produced. In addition, the reduction in sunlight associated with rainfall may result in reduced fiber maturity (Bradow and Bauer, 1997).

The micronaire of individual bolls varies by location of the cotton plant and is also influenced by the overall plant boll load. In general, higher micronaire bolls develop earlier in the flowering period and lower on the plant. Later bolls set higher on the plant, are typically lower in micronaire. In addition, a plant with a heavy boll load produces bolls with lower micronaire than a plant with a sparse fruit load (Heitholt, 1997).

All of these observations can be related to carbohydrate allocation. Bolls produced early in the season have an abundant carbohydrate supply via a large canopy of leaves relative to the number of bolls. Heat unit accumulation is also typically higher for early boll compared to late bolls. Later in the season bolls develop with a more mature leaf canopy, a preexisting boll load, and fewer heat units, all of which reduce the carbohydrate supply to later bolls. The same general relationship can be made for boll load on the plant. A plant possessing few bolls has many leaves available to supply carbohydrates, while a plant with many bolls has correspondingly fewer leaves available per boll to supply carbohydrates. Thus, any management factor which enhances boll load can reduce overall crop micronaire.

Defoliating a crop extremely early can reduce overall micronaire, but potentially at the cost of reduced yields. This reduction in micronaire is caused by the cessation of development of later bolls. However, maximum micronaire was achieved at 60 to 80 percent open boll in studies conducted in Georgia (Bednarz, et al., 2002). Thus, defoliating at the optimum time can influence micronaire but not to the extent of in-season management factors and boll load.

Harvest practices that allow for a once-over operation influence micronaire. Once-over harvesting blends higher micronaire bolls at the bottom of the canopy with lower micronaire bolls near the top of the canopy creating a crop micronaire that averages between the highs and lows.

Micronaire is different compared to other fiber parameters in that there is a base range, and deductions are assessed to bales possessing micronaire values above and below this range (USDA, 2001). There is also a premium range. The base range is from 3.5 to 4.9 with premiums assigned to micronaire readings ranging from 3.7 to 4.2 (Table 2). In Georgia micronaire problems are associated more often with high micronaire readings than low readings. These deductions for micronaire are also related to length. For example, a staple of 33 or longer with a micronaire of 5.0 would receive a deduction of $0.0365/lb, while a staple of 32 or shorter with a micronaire of 5.0 would receive a deduction of only $0.03/lb.

As discussed previously, micronaire is highly influenced by environment. Thus, there are no real discernable trends among southeastern locations over years, other than associations between hot and dry years. Locations incurring season-long drought or above-average temperatures will produce a crop with higher micronaire. Conversely, wetter and cooler production seasons will produce crops with lower micronaire.

Color Grade

Fiber color is most closely related to the growth environment at boll maturity and opening (Bradow and Davidonis, 2000). However, fiber color is also closely related to overall fiber quality because brighter and whiter fibers are of higher quality than dull gray or yellow fibers. Loss of brightness and whiteness is indicative of weathering and is often reflected in declines in fiber strength and length (Perkins, et al., 1984). Color grade is established via HVI analysis by determining the degree of reflectance (Rd) and yellowness (+b). Reflectance indicates a degree of brightness or dullness, and yellowness indicates the degree of color pigmentation. The color grade is then determined by locating the point at which Rd and +b values intersect on the Nickerson-Hunter cotton colorimeter diagram (Fig. 24) (USDA, 2001; Nickerson, 1950; ASTM, 1994).

Figure 24. Nickerson-Hunter cotton colorimeter diagram (http://www.ams.usda.gov/cotton/hvicolorupland.htm).

As the color of cotton deteriorates due to environmental conditions, the probability for reduced processing efficiency is increased. The deterioration of color also affects the ability of fibers to absorb and hold dyes and finishes (USDA, 2001).

otton defoliation and harvest timing are the management factors that can most significantly affect color grade. Timely defoliation and subsequent harvest can prevent loss of color grade due to weathering in the field (Bednarz, et al., 2002). A cotton boll is at its utmost quality the day that it opens.

Trash, Leaf Grade, and Extraneous Matter

The amount of non-lint materials in a cotton sample is considered trash (USDA, 2001). These materials may include leaf and bark from the cotton plant or weeds. HVI analysis determines trash content by scanning the surface of the cotton sample and calculating the percentage occupied by non-lint material (Fig. 25).

Figure 25. Color grade is determined by analyzing the bale sample in the above HVI instrument. The degree of reflectance and yellowing are measured from the sample ultimately delivering the color grade.

Leaf grade is a visual human estimate of the amount of plant leaf particles in the cotton sample made manually by a cotton classer (USDA, 2001). Leaf grade, extraneous matter, and preparation (discussed later) are the only classing parameters not determined by HVI. There are eight leaf grades, 1 through 8. Although trash and leaf grade are not the same, there is a correlation between the two as demonstrated in Table 3. The term "leaf" includes dried, broken plant foliage, bark and stem. These particles can also be classified into two general categories, large leaf and pin or pepper trash. Pin or pepper trash significantly lowers the value of the cotton to the manufacturer since it is more difficult to remove than larger pieces of trash (Perkins, et al., 1984; Moore, 1996; Xu et al., 1997).

The amount of leaf or trash remaining in the lint after ginning depends on the amount present in the cotton prior to ginning and the type and amount of cleaning and drying equipment used. From a manufacturers standpoint, leaf content is all waste, thus there is a cost factor associated with its removal (USDA, 2001). In addition, not all small particles can be removed from a bale. These small particles detract from the quality of the finished product. Leaf content in a bale can be affected by variety, harvesting methods and harvest conditions. A reduction in small leaf trash has been correlated with semi-smooth and super-okra leaf traits (Novick, et al., 1991).

Stems, burrs, bark, whole seeds, seed fragments, underdeveloped seeds (motes), grass, sand, oil, and dust are other types of trash found in lint samples. The environment influences the amount of wind-borne contaminants. Environmental factors related to pollination and seed development affect the occurrence of motes (Davidonis et al., 1996).

Extraneous matter can also be classified as any other type of trash that is not leaf or fiber. The kind of extraneous matter and an indication of the amount can also be noted by the classer on the classification document (USDA, 2001).

Preparation

Preparation is the term used to describe the degree of roughness or smoothness of ginned cotton lint (USDA, 2001). While various methods of harvest, handling, and ginning may produce differences in roughness of lint, no clear correlations have been found between preparation and spinning success. Figure (26) depicts a fiber sample exhibiting Level 1 preparation; one can note the twisted or ropey appearance of the fibers in the sample.

Figure 26. A bale sample exhibiting level 1 preparation. Note the knotted and twisted appearance of the sample. Preparation or lack thereof is visually documented by a cotton classer.

Other Factors Influencing Fiber Quality

Stink bugs and other plant bugs which feed on developing bolls can influence fiber quality as well as yield. These bugs feed on developing seeds within the boll and aid in the introduction of bacteria and fungi which degrade the lint in these bolls. In fact, the time at which stink bugs prefer to feed on bolls occurs when fibers are most rapidly elongating, and when fiber thickening is just beginning. Studies conducted in Georgia have shown that bolls on which stink bugs feed produce fibers that are lower in quality as indicated by almost all HVI measured parameters (Fig. 27).

Figure 27. Effect of stink bug infestation on various fiber quality parameters; A) Microniare, B) Staple, C) Uniformity, D) Strength, and E) Reflectance, on first position bolls by fruiting branch. Treated indicates weekly applications of Bidrin insecticide initiated at first flower. Data courtesy of Philip Roberts (previously unpublished, 2004). Yield in the untreated plots in this trial were reduced by 33%.

Current Roundup Ready® cotton technology allows for over-the-top applications of glyphosate to be made until the 5th true leaf is the size of a quarter. Over-the-top glyphosate treatments made beyond this 4 to 5 leaf stage are considered salvage applications and include specific warnings about the potential for reduction in yield. Sloppily-applied, direct treatments also pose a threat to yield. Yield reductions are common occurrences with such applications. However, studies to date have not indicated any detrimental effects on fiber quality (Culpepper et al., 2004).

Ginning of cotton could almost be considered a necessary evil in terms of preserving cotton fiber quality. The mechanical removal of the lint from the seed is necessary in order to make the lint useable to the textile industry. Proper ginning techniques also clean extraneous matter from the lint thus improving color grade. However, this process does damage some fibers and can potentially lower overall fiber quality. Gins across Georgia and the U.S. strive to strike a balance between efficient removal of fiber from the seed, lint cleaning, and quality preservation.

What Can Growers Do to Preserve Fiber Quality

The variety of cotton planted "sets the bar" for fiber quality. Variety selection, in a sense, determines the genetic potential for the quality of fiber to be produced. Length, strength, micronaire, and uniformity are all, some to a greater extent than others, genetically controlled. Once a commitment to a variety is made, the environment and growing conditions only reduce quality from its maximum genetic potential. Therefore, if producing for the utmost quality, the first step that must be taken is to select a variety with the best genetics for fiber quality.

The issue of balancing quality and yield is not as simple as it seems. Producers look for several attributes in a cotton variety. First and foremost is technology. Georgia farmers have rapidly adopted "stacked" varieties, those that contain the Bt endotoxin for worm pest control and Roundup Ready® technology together. Secondly, growers select varieties based on yield performance. Finally, growers select a variety with favorable fiber quality potential. At the current time a genetically engineered, high yielding, premium fiber quality variety is not available. In addition, current markets do not reward a producer for delivering a high quality fiber. Therefore, it does not make sense economically to sacrifice yield potential to produce higher quality fiber. Until the market shifts or varieties are available offering all three desirable traits, producers will most likely select varieties based on technology and yield potential. These trends make the following two items even more important in terms of producing a crop with high fiber quality.

The yield loss due to boll-feeding bugs has been well-documented. As discussed previously, bug feeding reduces fiber quality. Recent studies conducted in Georgia indicate that nearly all fiber quality parameters are reduced in bolls fed upon by stink bugs. Therefore there are multiple compelling reasons to control these pests.

Extensive work has been done in Georgia to indicate that timely defoliation and harvest are imperative for producing cotton of high quality. Georgia cotton producers commonly wait to apply harvest aids until 80 percent or more of the harvestable bolls are open. This practice, in conjunction with waiting several weeks after plants are completely defoliated to initiate harvesting, has been proven to significantly reduce fiber quality, yield and profit.

Studies conducted in Georgia from 1998 though 2000 consistently demonstrated that defoliants should be applied between 60 and 80 percent open boll and then the crop harvested within two weeks to optimize quality and yield. Delaying defoliant application and harvest beyond this time reduced fiber length, uniformity, strength, and lint yield (Bednarz et al., 2002). Furthermore, these studies also indicated that short fiber content (fibers less than 0.5 inches in length) increased as defoliant application and harvest was delayed. Delaying defoliant application and harvest past the optimal time reduced yield and quality. Net returns were reduced from $7 to $28 per acre per week.

Cotton is an indeterminate crop, thus it flowers and sets fruit over an extended period of time. This fruiting cycle may last as long as 6 weeks. Thus, once the plant has reached 60 percent open boll, the fiber in the first opened bolls may have been exposed to the weather for up to 3 weeks. As stated multiple times in this document, the fiber in a cotton boll is at its premium the day it opens. The environment reduces the quality of that boll from that day forward.

Conclusions

The cotton industry is rapidly moving toward a more quality driven market. The inherent problem is that the preponderance of cotton research has dealt with improving yields. Thus the research world is left to play catch-up with the industry. There will most likely continue to be a market for lower quality cotton but at a significant discount. Research, teaching, and extension personnel must continue to strive to identify not only yield enhancing practices but also those that improve quality.

Acknowledgements

The majority of the technical information for this bulletin was included in the paper entitled "Quantification of Fiber Quality and the Cotton Production-Processing Interface: A Physiologists Perspective," written by Judith Bradow and Gayle Davidonis, published in the 2000 issue of the Journal of Cotton Science (v4:p34-64). This paper provided an excellent review of current and past research relating to the physiological and processing aspects of cotton fiber quality.

Bulletin 1289 / July2005