Effects of cotton genotype,defoliation timing and season on fiber cross-sectional properties and yarn performance

Abstract

Cotton fiber cross-sectional properties influence the performance of ring spun yarns. The spinning performance of two Gossypium hirsutum L. Upland cotton genotypes known to have inherently different fiber fineness properties were compared. Genotypes were grown together in field experiments conducted over two growing seasons, and crops were subjected to early and late defoliation treatments. The aim was to quantify the differences in yarn properties following changes targeting fiber fineness properties in isolation from other fiber properties. For the first time, the percentage difference in yarn properties was captured along with the associated changes made to alternative fiber fineness properties within the base micronaire 3.50 to 4.90 G5 range. As expected the genotype with lower fiber micronaire, linear density, and perimeter, spun yarns that were stronger and more even. Late defoliated cotton plants produced fibers that were higher in micronaire and maturity ratio, and were bigger in perimeter, which demonstrated that the fibers had expanded during the secondary wall thickening phase of development. However, the defoliation treatment effect on fiber fineness properties was smaller compared with the effect of genotype, and no change to any yarn property was detected. In terms of environmental effects, the first season cotton had smaller perimeter finer fibers that spun stronger and more even yarns. In contrast, the second season cotton had bigger perimeter fibers that spun weaker and less even yarns.

Keywords

Cotton genotypes fiber quality fineness micronaire miniature spinning yarn performance

Most people utilize cotton fibers every day. As a fabric for apparel its versatility and superior next-to-skin comfort attributes keep it in demand. However, there is an ongoing need to enhance cotton’s competitiveness due to the increasing prevalence of manmade synthetic alternatives. Synthetic fibers are arguably cheaper to make, but are also a source of undesirable microplastic contamination of the environment. The degree of the fineness or coarseness of fiber is an important attribute that is critical to the performance of spun yarns and derived textile products. While synthetic fibers can be manufactured consistently with predetermined cross-sectional attributes like perimeter, shape, and linear density, cotton is much more variable because of its biological nature. Cotton plants are iteratively flowering perennials that produce bolls (fruit) of different maturities from the bottom to the top of the plants, so the timing of harvesting can also contribute to variation in fiber properties. Cotton fibers are long single cells that elongate from the surface of seeds during approximately the first 28 days post anthesis (DPA) of boll development. From approximately 17 DPA, a thick secondary cell wall of cellulose is deposited until a maturation period beginning at approximately 40 DPA.¹ When bolls are mature they open to expose fibers for harvest. Dried mature fibers are not uniformly round, but are shrunken, convoluted, irregularly shaped twisted ribbons due to their hollow lumens. These physical peculiarities provide a challenge for the effective and rapid measurement of cross-sectional properties to enable harvested cotton to be appropriately characterized and valued by producers and spinners.

The most popular, rapid, and still current method of measuring cotton cross-sectional properties for routine testing and valuation is the micronaire air resistance method. Micronaire is determined by both cotton fiber linear density,^2–4 which is a surrogate for fineness or perimeter, and maturity or the degree of secondary wall thickening (maturity ratio).^4,5 High micronaire cotton is considered coarse (large perimeter) by spinners and results in fewer fibers in the yarn cross-section, which translates into weaker, inferior yarns. Alternatively, while lower micronaire cotton is associated with more desirable finer, lower perimeter fibers, lower micronaire cotton may also occur because fibers are undesirably immature, which are prone to dye uptake problems, breakage, and entanglement (Figure 1).

Figure 1.

Microscope and analyzed images respectively of cotton fiber cross-sections as determined by a standard method.⁶ Cross-sections (a) and (b) represent cottons that have the same micronaire (Mic), but cotton (a) is less desirably coarser, being bigger in perimeter, (P), and higher in linear density, (H), but has less relative cell wall thickening with lower maturity ratio, (MR), compared with (b), the finer more mature and more desirable cotton. Cross-sections (c) and (d) represent finer, lower micronaire cottons. While both have the same micronaire, cotton (c) is coarser and bigger in perimeter and has a lower MR than (d). Adapted from Bange et al.⁷

Micronaire is widely adopted because it is based on fast, simple, and reliable mechanical technology, and is supported by international standards and universal calibration cottons. Other technologies have been developed that independently measure cotton fiber linear density and maturity, and therefore offer obvious advantages compared with measuring micronaire alone. There are various instruments and techniques,^8–14 but all are slower than micronaire and require varying degrees of extra skill to undertake testing. They also vary in the degree of directness in measuring either component, and there is a lack of universally accepted standards and, thus, discrepancies between methods. These factors govern where and when they are utilized; for example, whether they are used to independently retest cotton in commercial spinning mills to assist laydown management and processing, or by researchers attempting to breed finer cottons or to better manage crops to minimize the incidences of immature bolls and fibers.

Some of the factors that influence cotton cross-sectional properties are reasonably well-known. There is a clear genetic component with known differences occurring between cotton species. For example, Gossypium barbadense L. (Pima) cotton has finer, smaller perimeter fibers compared with commercially common Gossypium hirsutum L. (Upland) cotton, and cottons within a species can also vary in perimeter and fineness.^1,15 The indeterminant nature of cotton flowering means that plants need to be managed with chemical boll openers and leaf defoliants to ensure that the crop is prepared for a single machine harvest. Crop maturity management via defoliation timing therefore affects plant photosynthesis, cellulose deposition, and fiber maturity, and this can flow on to have an impact on post-harvest processing and textile performance.^16–18 Environmental conditions also play a role in determining fiber properties. For example, average daily temperature is known to affect the secondary wall thickening phase of fiber development, which can have an impact on micronaire.^19,20 While fiber perimeter is thought to be primarily genetically driven, there is conflicting evidence about the extent to which environmental conditions can affect perimeter, and whether, or by how much, fiber perimeter can change during the fiber development phases.^1,15,21

The aim of this study was to grow two cotton genotypes known to have inherently different fineness attributes, but that had similar other attributes such as length and strength. The intention was to isolate the influences that fiber cross-sectional fineness properties can have on the performance of 20 tex knit twist ring spun yarns. In addition to micronaire, alternative fiber measurements including linear density, maturity, perimeter, and width were also undertaken to assist in explaining the impacts of the experimental treatments, and to confirm and demonstrate the advantages of using cross-sectional measurements other than micronaire alone. New instrumentation that can measure these components more precisely will allow the cotton industry to look at these elements in new ways and establish more appropriate thresholds for making management decisions that have an impact on textile quality. One aim of the research was to utilize collected climate data to contribute to the understanding about how seasonal and environmental conditions have an impact on fiber and yarn properties. It was hypothesized that less mature fibers induced through an early defoliation treatment would affect a greater number of lower linear density fibers in the yarn cross-section and therefore affect yarn performance. It was also hypothesized that environmental conditions in different growing seasons could be demonstrated to influence fiber perimeter to the same degree as a genotype effect, or to a significant enough degree to subsequently have an impact on yarn performance. While other studies have attempted to capture similar outcomes, these have mostly focused on using only micronaire as the measurement of fiber fineness.^22–24 Further, there are still unexplained fiber components that contribute to yarn properties.²⁵ An attempt to capture the percentage change in yarn performance attributes following a targeted change made to cotton fiber cross-sectional fineness properties other than just micronaire, and using genotype, management, and season to induce the changes, is research not known to have been published previously. Certainly, such research, specifically using Australian-bred genotypes grown under Australian conditions, has not been undertaken previously.

Methods and materials

Cotton production

Two field experiments (Exp.) were conducted over two consecutive seasons (Exp. 1 – the 2010/2011 season; and Exp. 2 – the 2011/2012 season) at the Australian Cotton Research Institute (ACRI) near Narrabri (30°19’0”S, 149°46’0”E) in north-western New South Wales, Australia. This is a semi-arid environment with a uniform, gray, cracking, clay soil type (USDA Soil Taxonomy: Typic Haplustert). The two experiments examined two Australian Upland cotton (G. hirsutum L.) genotypes, both experimental lines from the Commonwealth Scientific and Industrial Research Organisation (CSIRO) cotton breeding program. One genotype (981-117) was known to have higher fiber micronaire and higher fiber linear density than the other (981-41). Therefore, theoretically they were inherently different from each other from a fiber perimeter or fineness standpoint. Both genotypes had similar other fiber properties like length and strength. Genotypes were examined under two defoliation scenarios, an early treatment implemented when approximately 20% of bolls were open, and a control treatment when approximately all bolls (100%) had opened. Experiments were set up as a randomized complete block design with four replications.

Experiments were sown with a commercial row crop planter (Exp. 1–21 October 2010; Exp. 2–17 October 2011) with a between-row spacing of 1 m (Table 1).

Table 1.

Dates and corresponding number of days after sowing (DAS) for key crop production events for the two field experiments. Two defoliation treatments were implemented in each experiment: Early at approximately 20% open bolls (OB) and control at approximately 100% OB.

	Exp. 1		Exp. 2
	Date	DAS	Date	DAS
Sowing	21/Oct/2010	0	17/Oct/2011	0
Defoliation
Early 20% OB	8/April/2011	169	30/March/2012	165
Control 100% OB	20/April/2011	181	21/April/2012	187
	29/April/2011	190	29/April/2012	195
Harvest	13/May/2011	204	14/May/2012	210

Treatment plots were four rows wide for Exp. 1 and eight rows wide for Exp. 2. Plots were initially planted at 9 m in length, which were chipped back to 7 m after plants were established. Experiments were established and grown using normal Australian production practices, including full furrow irrigation, the use of non-limiting nitrogen, and using standard pest and weed controls.^26,27 Nitrogen was applied as anhydrous ammonia (injected below and to the side of the plant line) at least 1 month before sowing at a rate of 180 kg/ha.

In a fixed area of 1 m² in each control plot from the time that bolls had begun to develop, a count of all cotton bolls was made and the number of open bolls (defined as from when two sutures on the boll had split) was also noted. The percentage of open bolls was calculated as:

% open bolls = \frac{Number of open bolls}{Number of all bolls} \times 100

(1)

Early defoliation treatment plots were sprayed once with a calibrated CO₂ pressurized boom sprayer with a total swath width of 3 m using flat fan nozzles (110-01) at 200 kPa delivering 100 L/ha of spray solution. The entire field including the early and control defoliation treatments were sprayed again twice by plane when the control plots had reached 100% open bolls (Table 1). The chemical and rates for both defoliation treatments were: 0.2 L ha⁻¹ Dropp Liquid® (Bayer CropScience, active constituent 500 g/L Thidiazuron); 3 L/ha Prep 720® (Bayer CropScience, active constituent 720 g/L Ethephon); and 2 L/ha D-C Tron® (Caltex, active constituent 991 ml/L Petroleum Oil).

Meteorological data were recorded over the course of each experiment via a weather station located at ACRI. Accumulative day degrees were calculated for both experiments using maximum air temperature (Tmax) and minimum air temperature (Tmin) data:²⁸

Day degrees = \frac{(Tmax - 12) + (Tmin - 12)}{2}

(2)

When minimum temperatures were less than 12°C:

Day degrees = \frac{Tmax - 12}{2}

(3)

Harvesting and ginning

One central row of cotton from each treatment plot was harvested with a modified spindle basket cotton picker (John Deere, Moline, Illinois, USA) that had a single harvester head, and the capacity to capture and tag smaller amounts of experimental cotton. One sub-sample of approximately 1 kg of seed cotton was taken from each plot and ginned to separate cotton fiber from seed using a 20 saw gin (Continental Eagle, Prattville, Alabama, USA). No lint cleaning was undertaken on ginned fiber, to minimize any confounding influences on fiber breakage from any mechanically intensive lint cleaning. Ginning produced approximately 400 g of fiber per experimental plot (32 in total from both experiments), which was subsequently sub-sampled for fiber testing and yarn production.

Fiber quality measurements

Approximately 50 g of ginned fiber for each plot was subjected to testing using an Uster High Volume Instrument (HVI) model 1000 (Uster Technologies AG, Sonnenbergstrasse, Switzerland).²⁹ HVI properties are used universally to grade and value cotton.³⁰ They assist spinners in laydown preparation and yarn processing, and relate well to and assist in forecasting yarn performance.^31,32 HVI properties that were measured for the study herein were micronaire, upper half mean length (mm), which is the average length of the longer half of fibers by weight; length uniformity (%), which is the ratio of the mean length to the upper half mean length; short fiber index (%), which is an estimate of the percentage of fibers that are shorter than 12.7 mm; strength (g/tex); and elongation (%). HVI results per plot were the average of two replicate tests made within each 50 g sample.

Approximately 45 g of ginned fiber per plot was subjected to one pass of a Shirley Analyzer (SDL, Stockport, UK) to remove vegetative and extraneous matter. Some 10 to 15 g of this cleaned fiber was used to prepare 50 mg lots of 0.7 mm long snippets made with a custom-built hand guillotine. These snippets were tested in the Cottonscope instrument to determine fiber linear density (mtex).⁸ Cottonscope results were the average of three replicate tests per sample.

The external geometry of cotton fibers is influenced by the combination of perimeter and maturity, with parameters like fiber circularity and fullness,^33,34 and the diameter or width of fibers³⁵ also being acknowledged to represent an indication of cotton fineness. Therefore additional 15 g sub-samples of cleaned fiber were subjected to an air-driven piston coring instrument³⁶ to generate approximately 300 mg lots of 2 mm long fiber snippets. These snippets were subjected to fiber width (µm) determination using the CSIRO laser photometric instrument Sirolan Laserscan.^37–40 Results for fiber width determination were the average of three replicate tests per sample.

Based on the method described by Hequet et al.,⁶ 15 g sub-samples of cleaned fiber were also subjected to direct microscopic cross-sectional analysis (examples in Figure 1). This is arguably the most definitive, direct, and sensitive method of determining cross-sectional parameters, but it is also the most time-consuming and requires specialist skills.^41,42 Briefly, combed fiber samples were embedded in a methacrylate polymer and 1 μm thick fiber cross-sections were obtained using a microtome. Slide-mounted sections were observed with a microscope and the captured digital images were analyzed by custom image analysis software to measure fiber perimeter, cross-sectional area, and lumen area; cell wall area is cross-sectional area minus lumen area. Fiber maturity in this case was a direct determination of the degree of secondary wall thickening (or theta, θ), which is the ratio of the cell wall area to the area of a circle having the same perimeter as the fiber of interest, and is directly related to maturity ratio (equation 4).⁴³ Fiber maturity ratio and perimeter results were determined by this method. Approximately 450 single fiber cross-sections were analyzed per sample.

Maturity ratio = \frac{θ}{0.577}

(4)

Ring spun yarn manufacturing

Sub-samples of ginned cotton fiber were cleaned and prepared for spinning using a small sample cotton processing plant manufactured by Tianjin Jiacheng Mechatronic Equipment Co. Ltd. (Tianjin, China). The process was as follows: 130 g of ginned cotton was sampled from each plot sample and subjected to one pass of a DSOP-02 sample opening machine to remove large trash particles. From this opened cotton, six 20 g portions were then each carded twice via a DSCa-01 carding machine. Each of the six produced carded batts were then separately drawn and formed into six slivers using a DSDr-01 drawing machine, and these were subsequently blended to form a single sliver via a second pass of the drawing machine. The single drawn sliver was then converted into twisted roving using a DSRo-11 roving machine.

Bobbins of twisted roving were spun into yarns via a sample six spindle Ser. Ma. Tes. Spinntester ring spinning machine (Ser. ma. tes. srl, Cologne, Italy). Yarns spun were 20 tex (g/1000 m) (or 30 imperial English cotton count) with a knitting twist (αe 3.7). These parameters were determined using standard in mill procedures and equipment, and calculations utilizing measured spindle and roller speeds. One to two bobbins of yarn were produced per experimental plot.

Yarn testing

All yarn testing was conducted under standard textile testing air conditions (20°C ± 2°C and 65% RH ± 2%). Each of the yarn tests were sequentially completed once, and then the testing procedure was repeated another two times. Yarn results per experimental plot were the average of the three replicate tests made.

Yarn linear density or count (tex) was determined by weighing 100 m long portions of yarn following oven drying and adequate laboratory conditioning, and yarn twist (turns per meter) was determined using a Zweigle yarn twist machine.

Yarn mass and geometrical variation parameters were determined using an Uster Tester 4-SX instrument (Uster Technologies AG, Sonnenbergstrasse, Switzerland). Reported parameters were average yarn diameter (mm); evenness [% coefficient of variation (CVm %)], which gauges the variation in the mass of multiple 0.01 m portions of yarn; thick places (+50%), which is the number of places identified as being greater than half the thickness of the yarn per 1000 m; thin places (−50%), which is the number of places identified as being less than half the thickness of the yarn per 1000 m; and hairiness (H), which corresponds to the total length of protruding fibers (in cm) within the measurement field of 0.01 m.

An estimation of the average number of fibers in the yarn cross-section was calculated as the average linear density or count of the yarn in tex divided by the average linear density of the fibers in tex used to manufacture the yarn,

No . fibers in yarn cross section = \frac{Yarn count in tex}{fiber count in tex}

(5)

An estimation of the packing density of the number of fibers per square millimeter of yarn cross-section was calculated as

Packing density = \frac{Number of fibers in yarn cross section}{3.1416 × {(\frac{Yarn diameter}{2})}^{2}}

(6)

An estimation of the density of yarn in grams per cubic centimeter was calculated as

Yarn density = \frac{Yarn tex}{3.1416 × {(\frac{Yarn diameter}{2})}^{2} \times 100000}

(7)

Yarn tensile properties were determined using an Uster Tensorapid 3 instrument (Zellweger Uster, Sonnenbergstrasse, Switzerland). Reported tensile properties were force to break (cN); tenacity (cN/tex), which is the force to break divided by the tex of the yarn; elongation (%); and work to break (cN.cm), which can be visualized as the area under the stress strain curve.

Statistical analysis

Microsoft Excel was used for basic data handling. The General Analysis of Variance facility in Genstat 19 (Lawes Agricultural Trust, IACR, Rothamsted, UK) was used for analysis of variance (ANOVA) of data. To test the effects of genotype and defoliation treatments on fiber and yarn attributes, both experimental years were combined in a single analysis model, with year being designated a random term. It was a two-factor analysis. The treatment structure was genotype × defoliation with a block structure of year/block. No significant interactions were captured between the genotype and defoliation variables for all fiber and yarn attributes, therefore main effects mean values were reported along with their relevant degrees of statistical significance. In a separate statistical exercise, one-factor ANOVA was employed to test the effect of experimental year on fiber and yarn attribute results. An indication of the significance of ANOVA results was documented by standard abbreviations *p < 0.05, **p < 0.01, ***p < 0.001, and NS: not significant.

Where there was a significant effect of treatments, the percentage difference between relevant mean values was calculated as

% difference = \frac{| A - B |}{(\frac{A + B}{2})} \times 100

(8)

where A and B are the two mean values being compared.

Results

Effects of genotype and defoliation timing

Fiber quality

Genotype 981-117 fibers were coarser and more mature, with this genotype having significantly higher fiber micronaire (by 0.36 micronaire or a 9.3% difference), maturity ratio (by 0.04 or a 5.1% difference), linear density (20.5 mtex or an 11.8% difference), perimeter (2.19 µm or a 4.0% difference), and width (0.61µm or a 4.1% difference) compared with genotype 981-41 (Table 2). There was no difference between genotypes for upper half mean length, although 981-117 cotton was lower in length uniformity and higher in short fiber content compared with 981-41 (both significant at p < 0.05). There was no difference between genotypes for fiber bundle strength and elongation (Table 2).

Table 2.

Fiber and yarn performance results for the main effects of genotype and defoliation timing treatments. Values are mean ± standard deviation, and relevant treatment ANOVA comparison statistics *p < 0.05, **p < 0.01, ***p < 0.001, and NS: not significant. Percent difference (% diff) values are reported for statistically significant results. There were no significant interactions.

	Genotype				Defoliation
Attribute	981-117	981-41	p	% diff	Early	Control	p	% diff
Fiber quality
Micronaire	4.07 ± 0.46	3.71 ± 0.53	***	9.3	3.76 ± 0.46	4.02 ± 0.56	**	6.7
Maturity ratio	0.81 ± 0.04	0.77 ± 0.04	***	5.1	0.78 ± 0.04	0.80 ± 0.05	**	2.5
Linear density (mtex)	184.6 ± 21.9	164.1 ± 23.5	***	11.8	171.3 ± 21.6	177.4 ± 27.7	NS	–
Perimeter (µm)	55.57 ± 1.25	53.38 ± 1.56	***	4.0	53.83 ± 1.66	55.12 ± 1.83	**	2.4
Width (µm)	15.18 ± 0.34	14.57 ± 0.16	***	4.1	14.88 ± 0.40	14.88 ± 0.42	NS	–
Upper half mean length (mm)	30.47 ± 0.84	30.87 ± 0.91	NS	–	30.83 ± 0.99	30.51 ± 0.76	NS	–
Length uniformity (%)	81.21 ± 1.86	82.48 ± 1.53	*	1.6	82.11 ± 1.46	81.58 ± 2.09	NS	–
Short fiber content (%)	11.36 ± 2.24	9.68 ± 2.01	*	16.0	10.25 ± 2.31	10.79 ± 2.25	NS	–
Strength (g/tex)	30.13 ± 1.14	29.49 ± 0.83	NS	–	29.78 ± 0.97	29.84 ± 1.13	NS	–
Elongation (%)	6.53 ± 0.64	6.74 ± 0.85	NS	–	6.62 ± 0.72	6.64 ± 0.81	NS	–
Yarn quality
Count (tex)	18.69 ± 0.59	19.56 ± 1.20	*	4.5	18.90 ± 0.63	19.34 ± 1.30	NS	–
Diameter (mm)	0.242 ± 0.005	0.240 ± 0.004	NS	–	0.242 ± 0.004	0.241 ± 0.005	NS	–
Twist (turns/m)	843.8 ± 49.3	840.9 ± 49.3	NS	–	845.7 ± 49.0	839.0 ± 49.5	*	0.8
Fibers in cross-section	102.6 ± 12.9	121.6 ± 19.2	***	16.9	112.1 ± 15.5	112.1 ± 22.1	NS	–
Packing density (fibers/mm²)	2230 ± 253	2679 ± 389	***	18.3	2448 ± 357	2461 ± 442	NS	–
Density (g/cm³)	0.407 ± 0.014	0.432 ± 0.029	***	6.0	0.413 ± 0.018	0.426 ± 0.030	NS	–
Evenness (CVm %)	21.16 ± 1.88	18.55 ± 1.28	***	13.1	19.59 ± 2.04	20.13 ± 2.12	NS	–
Thick (+50%)	725 ± 296	374 ± 174	***	63.9	509 ± 271	590 ± 326	NS	–
Thin (–50%)	372 ± 232	136 ± 96	***	92.9	227 ± 179	281 ± 243	NS	–
Hairiness (cm)	6.05 ± 0.30	5.55 ± 0.41	***	8.6	5.75 ± 0.46	5.84 ± 0.42	NS	–
Breaking force (cN)	252.5 ± 25.9	288.1 ± 33.8	***	13.2	267.2 ± 30.6	273.4 ± 39.1	NS	–
Tenacity (cN/tex)	13.53 ± 1.37	14.76 ± 1.52	***	8.7	14.16 ± 1.57	14.12 ± 1.59	NS	–
Elongation (%)	7.06 ± 0.57	7.66 ± 0.35	***	8.2	7.43 ± 0.58	7.29 ± 0.54	NS	–
Work to break (cN.cm)	401.6 ± 39.5	492.4 ± 74.5	***	20.3	445.2 ± 69.9	448.7 ± 81.4	NS	–

For the effect of defoliation, fiber from the control defoliation treated plants had statistically significantly higher micronaire by 0.26 (or a 6.7% difference), maturity ratio by 0.02 (or a 2.5% difference), and perimeter by 1.29 µm (or a 2.4% difference) compared with early defoliated cotton (Table 2). The defoliation treatment did not affect any other fiber quality attribute.

Yarn quality

Yarn count for genotype 981-41 was marginally but significantly higher by 0.87 tex compared with 981-117. There was no difference between the two genotypes for either yarn diameter or twist. Yarn twist for the control defoliation timing cotton was lower (by 6.7 turns/m or a 0.8% difference) compared with the early defoliation treated plants, with there being no other significant effect of defoliation treatments on the other yarn attributes (Table 2). Genotype 981-41 yarns had on average 19 more fibers in the yarn cross-section (16.9% difference), were higher in packing density (by 449 fibers/mm² or 18.3%) and were greater in density (by 0.025 g/cm³ or a 6.0% difference) compared with 981-117 yarns. Genotype 981-41 yarns had lower evenness (13.1% difference), had 351 fewer thick places (63.4% difference), had 236 fewer thin places (or a 92.9% difference), and were less hairy (8.6% difference) compared with genotype 981-117. For yarn tensile properties, genotype 981-41 yarns were stronger with higher breaking force (by 35.6 cN or a 13.2% difference) and higher tenacity (by 1.23 cN/tex or 8.7%) compared with 981-117 yarns (Table 2). For yarn toughness, 981-41 yarns had higher elongation by 0.6% (or an 8.2% difference) and higher work to break by 90.8 cN.cm (a 20.3% difference) compared with 981-117.

Seasonal effects

Weather data

Both experiments initially had similar average air temperatures around sowing and early seedling growth, following which Exp. 2 recorded higher temperatures during November. During the central four months of production, higher air temperatures were recorded for Exp. 1 (Figure 2(a)).

Figure 2.

(a) Average monthly air temperature, solar radiation, and (b) total monthly rainfall, for both experiments.

During Exp. 2 there was a marked spike in air temperature in mid-November at approximately 30 days after sowing (DAS), and the hottest days recorded occurred in late January in Exp. 1 (Figure 3).

Figure 3.

Average daily temperature (spikey lines) and accumulative day degrees (smooth lines) results for the periods of cotton production for both experiments.

These results were reflected in cumulative day degree calculations, which initially tracked higher for Exp. 2 during vegetative growth, were similar for both experiments between 75 and 90 DAS during peak flowering, then tracked higher for Exp. 1 from 95 DAS during boll development and maturation (Figure 3). For average solar radiation, Exp. 1 had lower solar radiation for the first 3 months of production, and then more solar radiation for the following 2 months (January and February), with average monthly levels alternating for the rest of production (Figure 2(a)). Rainfall was similar for both experiments, except for markedly more being measured during January and February in Exp. 2, and more during March in Exp. 1 (Figure 2(b)).

Fiber quality

In comparing both experiments, on average fiber from Exp. 1 was finer and had less cell wall thickening, with Exp. 1 fibers having significantly lower micronaire (by 0.78 micronaire or a 20.1% difference), maturity ratio (by 0.07 maturity ratio or an 8.8% difference), linear density (by 37.5 mtex or a 21.5% difference), perimeter (by 2.03 µm or a 3.7% difference), and width (by 0.42 µm or a 2.8% difference) (Table 3).

Table 3.

Fiber and yarn performance results for each experiment. Values are mean ± standard deviation, and relevant between experiment ANOVA comparison statistics *p < 0.05, **p < 0.01, ***p < 0.001, and NS = not significant. Percent difference (% diff) values are reported for statistically significant results.

Attribute	Exp. 1	Exp. 2	p	% diff
Fiber quality
Micronaire	3.50 ± 0.40	4.28 ± 0.28	***	20.1
Maturity ratio	0.76 ± 0.03	0.83 ± 0.03	***	8.8
Linear density (mtex)	155.6 ± 18.04	193.1 ± 13.40	***	21.5
Perimeter (µm)	53.46 ± 1.60	55.49 ± 1.39	**	3.7
Width (µm)	14.67 ± 0.25	15.09 ± 0.42	**	2.8
Upper half mean length (mm)	31.02 ± 1.00	30.32 ± 0.59	*	2.3
Length uniformity (%)	81.1 ± 2.0	82.6 ± 1.2	*	1.8
Short fiber content (%)	10.28 ± 2.84	10.76 ± 1.53	NS	–
Strength (g/tex)	29.94 ± 0.94	29.68 ± 1.13	NS	–
Elongation (%)	5.98 ± 0.27	7.29 ± 0.42	***	19.7
Yarn quality
Count (tex)	19.15 ± 1.34	19.10 ± 0.62	NS	–
Diameter (mm)	0.242 ± 0.005	0.240 ± 0.005	NS	–
Twist (turns/m)	795.5 ± 8.57	889.2 ± 9.80	***	11.1
Fibers in cross-section	124.8 ± 17.7	99.4 ± 8.4	***	22.7
Packing density (fibers/mm²)	2708 ± 376	2201 ± 210	***	20.7
Density (g/cm³)	0.416 ± 0.029	0.423 ± 0.022	NS	–
Evenness (CVm %)	18.64 ± 1.19	21.08 ± 2.05	***	12.3
Thick (+50%)	373 ± 151	726 ± 307	***	64.2
Thin (–50%)	142 ± 73	366 ± 246	**	88.2
Hairiness (cm)	6.04 ± 0.36	5.55 ± 0.36	***	8.5
Breaking force (cN)	294.4 ± 29.1	246.2 ± 20.0	***	17.8
Tenacity (cN/tex)	15.37 ± 0.89	12.92 ± 0.99	***	17.3
Elongation (%)	7.03 ± 0.56	7.69 ± 0.31	***	9.0
Work to break (cN.cm)	492 ± 77	402 ± 36	***	20.1

For length, Exp. 1 fiber was marginally longer (by 0.7 mm or a 2.3% difference) but had lower length uniformity by 1.5% (1.8% difference). Both experiments had fiber with the same short fiber content. For fiber tensile properties, there was no difference between the two experiments for bundle strength, but Exp. 2 fibers were higher in elongation by 1.31% (19.7% difference) (Table 3).

Yarn quality

For yarn quality between experiments, on average Exp. 1 yarns had 93.7 turns/m less twist (11.1% difference), had 25.4 more fibers in the cross-section (22.7% difference) and higher packing density by 507 fibers/mm² (20.7% difference) compared with Exp. 2 yarns (Table 3). Exp. 1 yarns on average were significantly more even by 2.44 CVm % (12.3% difference), had 353 fewer thick places (64.2% difference), had 224 fewer thin places (88.2% difference), and were hairier by 0.49 cm (8.5% difference). For tensile properties, Exp. 1 yarns on average were stronger with higher breaking force (by 48.2 cN or a 17.8% difference) and higher tenacity (by 2.45 cN/tex or a 17.3% difference). Exp. 1 yarns were lower in elongation by 0.66% (9% difference) compared with Exp. 2 yarns. Exp. 1 yarns were tougher having higher work to break by 90 cN.cm (or a 20.1% difference) (Table 3).

Discussion

Effects of genotype and defoliation timing

Fiber for both genotypes had reasonable but not high degrees of secondary wall thickening. Depending on the classification system used, the genotypes were either designated as immature, or 981-41was immature (<0.80 maturity ratio) while 981-117 was classified as mature.⁴⁴ This highlights and confirms the inaccuracies of micronaire, because regardless of their level of fiber maturity, both genotypes were in the premium micronaire range (3.70 to 4.20).⁴⁵ Cotton in this range usually receives a 10 US cent per pound (or “points”) premium, and it sits within the base G5 micronaire range of 3.50 to 4.90, which receives neither a price premium nor a penalty. The genotype maturity ratio difference captured here will have physically influenced the micronaire, but certainly a greater effect would have been that of the different fiber perimeters and the subsequent impact this had on fiber linear density and yarn performance attributes. While 981-41 had marginally less mature fibers, the perimeter-driven lower fineness and linear density and therefore lower micronaire of this genotype contributed to yarns with more fibers in the cross-section, that were higher in density, and that had fewer thick and thin places. This made for yarns that were more even and that had fewer weak links,⁴⁶ and that therefore were stronger and tougher with higher tenacity and higher work to break results compared with 981-117 yarns. Of course, other fiber properties will affect yarn performance, and while tensile and average length results for both genotypes were similar, it is conceded that the small but significant differences in length uniformity and short fiber content would have had a positive impact on yarn properties like evenness.⁴⁷ Certainly, the lower short fiber content of 981-41 would potentially have made for a reduction in the incidence of fiber ends protruding from the surface of the yarns, and the reason for the lower yarn hairiness for that genotype.

In comparison to other published research, Faerber⁴⁸ reported the fiber quality and open end spun yarn performance of several Australian Upland genotypes, and warned of the unreliability of using micronaire alone. In that study, two of the genotypes (Siokra V16 and Sicot 189) had micronaire of 3.8 and 4.3 respectively, and both fell into the premium micronaire range. Fiber fineness was 156 and 179 mtex, and yarn tenacity was 13.1 and 11.4 cN/tex (13.9% difference) respectively. The other two genotypes (Sicala 40 and Sicala V2i) had similar micronaire (4.2 and 4.1 respectively), but in this case the higher micronaire cotton was finer and more mature, which spun significantly stronger and more even yarns. In other related work, Ramey et al.⁴⁹ reported that fiber maturity was correlated well with yarn tenacity, and May and Taylor⁵⁰ reported the benefits of using fiber fineness attributes in conjunction with length and strength to improve the selection of cottons with improved yarn tenacity.

For the effect of defoliation timing, the early treatment prematurely halted plant function and reduced the deposition of cellulose in fibers. It also stopped the expansion of fibers, with the later treatment having bigger perimeter fibers. Average fiber width was no different between the two defoliation treatments. As fibers matured and filled with cellulose, they will have become more circular leading to a reduction in average width. In this case the fiber maturity ratio difference between defoliation treatments was approximately half that of the genotype effect, and the difference in perimeter was also less than the genotype effect, and this was reflected in a relatively smaller but statistically significant impact on micronaire. However, these differences were not enough to influence fiber linear density. Therefore, there was no measurable impact on the number of fibers in the cross-sections of yarns or to yarn density measures, and no impact on yarn tensile and evenness properties.

There is conflicting evidence in the literature that the perimeter of fibers can change during development. Factors that affect micronaire such as climate, carbohydrate supply dynamics, and boll retention and management, are usually explained by how they have an impact on relative cell wall thickening.⁵¹ But these factors are likely to interact with different genotypes to influence the degree to which fiber perimeter will change during development. This has implications for the understanding of what affects cotton fiber cross-sectional properties, and how, for example, decision support models are developed to manage crop micronaire. Seagull et al.¹⁵ employed a direct manual microscopic measurement technique and reported that fiber diameter continued to increase throughout both the early lengthening phase of fiber development, and during the later thickening phase. In the Seagull paper, an average change in fiber diameter during the final 20 days of development of 3.1% for three Upland genotypes was measured. There was some variation between them, with one Upland genotype changing very little (by 0.5 µm or 1.0% during the final 20 days), while the Pima genotype changed the most (2.2 µm or 4.5% change). In comparison to the results reported in this manuscript, an equivalent average 2.4% difference in increasing fiber perimeter during the final 17 days of development (the time between the early and the control defoliation) was measured. Petkar et al.²¹ reported microscopic cross-sectional properties of developing fibers for four species of cotton from 28 to 63 DPA. The Petkar et al.²¹ work showed that, as expected, fiber circularity and secondary wall thickness increased during development, but fiber perimeter did not change. The sectioning and measurement methods (projection microscope and manual tracing) were different to what was used for the work in this manuscript, and the populations had markedly larger standard deviations (approx. 7 to 10 times) and smaller sample sizes (n = 200). Chauhan et al.⁵² also measured the cross-sectional properties of various cotton genotypes using a similar method to Petkar et al.²¹ (and the same sample number) at 45, 55, and 65 DPA stages of development. The Chauhan et al.⁵² results were similar, except that increases in fiber perimeter with increasing developmental age were detected. Those authors suggested that changes in perimeter during fiber development were dependent on the opposing factors of increasing maturity and increasing “coarseness,” and that a propensity for greater maturity tends to go with lower perimeter fibers. Hessler⁵³ hand harvested open cotton bolls from flowers set before or after a midway point in the flowering window, and assessed the quality of this early and late cotton for each of 18 genotypes. The Hessler paper reported that cotton fibers from early set bolls were more mature, had smaller perimeters, and higher micronaire compared with later set bolls.

Seasonal effects

On average, cotton from both experimental seasons fell into the base G5 micronaire range. But the higher micronaire cotton from Exp. 2 arguably sat in the 10-point premium range (at two decimal places of precision for micronaire as measured by the HVI 1000 instrument), due to its marginally higher maturity ratio and greater perimeter and width. In this case, on average, Exp. 1 cotton yarns were stronger, more even, and performed better because they were made up of smaller perimeter finer fibers that made for yarns with more fibers in their cross-sections. Indeed, this work demonstrated that seasonal or environmental influences can affect cross-sectional properties and yarn performance to a similar or a greater degree as a genotype effect. For example, the average difference between experiments for fiber linear density, the number of fibers in the yarn cross-section, and yarn tenacity was 21.5%, 22.7%, and 17.3% respectively (Table 3), compared with the average genotype difference of 11.8%, 16.9%, and 8.7% respectively for the same fiber and yarn performance attributes (Table 2). While there were also unintended differences captured between experiments for other fiber properties like fiber length, these were relatively small occurring at a lower percent difference and at the lowest level of statistical significance compared with the cross-sectional properties. While those properties will have influenced yarn performance, a lesser role was played by them than cross-sectional properties.

Evaluating the climate data for the two experiments assists in explaining the observed fiber quality differences. Bange et al.¹⁹ presented models for micronaire based on average daily air temperature during an estimated fiber thickening period. That research approximated the start of flowering to be at 777 day degrees, and the point at which all bolls had reached the fiber thickening phase to be at 1437 day degrees. Therefore, using these calculations as a reference, the start of flowering for the work reported herein was estimated to have occurred at 75 DAS, which coincided at the start of January for both experiments. Similarly, the time that all bolls on plants had begun fiber thickening was estimated to have occurred at the end of January at 100 DAS and 110 DAS for Exp. 1 and Exp. 2 respectively (Figure 3). Presumably, for most of January flowering was occurring, which coincided with the period with the biggest differences in average daily temperatures, with Exp. 1 having higher daily temperatures than Exp. 2 (Figures 2 and 3). A considerable difference in temperature remained into February and March during the fiber filling phase of development. This time also coincided with more rainfall and shady conditions for Exp. 2. Goynes et al.⁵⁴ showed that light and temperature related to season affected cellulose deposition in fibers, and Pettigrew⁵⁵ demonstrated that shady conditions can influence cotton fiber quality by reducing photosynthesis. Lokhande and Reddy²⁰ measured fiber quality properties for cotton gown under different temperatures from flowering and reported that micronaire increased with increasing average daily temperature to 26°C but then declined at higher temperatures. The Lokhande and Reddy work also showed that the AFIS instrument-derived maturity ratio increased linearly with increasing temperature, which suggests that the decrease in micronaire was due to the higher temperatures producing smaller perimeter fibers. Certainly, the results of lower fiber micronaire and perimeter on average for Exp. 1 can be attributed to these same effects. Conversely, Pettigrew⁵⁶ grew two Upland genotypes at both control ambient and at a higher canopy air temperature by approx. 1°C warmer. Average temperatures ranged between 25.2°C and 27.8°C. It was reported that the warmer temperature treatment produced fiber with high maturity, but no difference was detected for fiber perimeter between treatments. It can be speculated that these environmental conditions potentially influenced fiber perimeter determination at early flowering by directing the number, or the size of, seed coat cells differentiating into elongating fibers. For example, Xie et al.⁵⁷ reported that field temperature during fiber initiation and early elongation profoundly influenced final fiber length, so it is possible that cell size attributes like perimeter could be similarly affected. The slightly shorter fibers for Exp. 2 can be also be attributed to the shady conditions for that season reducing the assimilate supply to fibers during elongation.

Summary and conclusions

This research enabled the assessment of some of the impacts that genotype, agronomic management, and season can have on cotton fiber cross-sectional properties and how they subsequently affect yarn performance. The work utilized unreleased Australian cotton genotypes. It also employed non-traditional measures of cotton fiber width and perimeter, and used miniature sample spinning technology to manufacture ring spun yarns for testing. This research provides new knowledge regarding how genotype, season, the environment, and defoliation practice can have an impact on cotton fiber fineness properties in relative isolation from other fiber properties. Specifically, it reports for the first time the percentage change in fiber cross-sectional fineness attributes, and the associated percentage changes made to 20 tex ring spun knit twist yarn performance properties.

The two experimental cotton genotypes produced fiber that on average fell within the premium 3.70 to 4.20 micronaire valuation grade. One genotype was finer having fibers that were 8.8% lower in micronaire, 11.1% lower in linear density, and were smaller in perimeter by 3.9% compared with the coarser genotype. These differences translated into the finer genotype producing yarns that were 9.1% higher in tenacity and that were 14.1% more even than coarser genotype yarns. Later control defoliated cotton produced fiber that was 6.9% higher in micronaire due to more cell wall thickening and cell expansion. It was therefore 2.6% higher in maturity ratio, and fibers were 2.4% bigger in perimeter compared with early defoliated cotton. However, the defoliation treatment did not affect fiber width or influence the number of fibers in the yarn cross-section, and no impact on yarn performance was measured. Seasonal influences on fiber cross-sectional properties were at least as much as the genotype effects. Cotton fibers from the first season were, on average, 18.2% lower in micronaire, 19.4% lower in linear density, and 3.7% smaller in perimeter compared with second season cotton. These differences bestowed a detectable impact on yarn quality, with first season yarns having more fibers in their cross-sections, which were 19% higher in tenacity and 11.6% more even than second season yarns.

These results demonstrate clear genetic determinants on the degree of fiber fineness or coarseness. As expected, they also demonstrated that early defoliation produces less mature fibers but, in this case, the effects on cross-sectional fineness and average fiber width were much smaller than seen with either genotype or seasonal effects. The evidence that cotton fibers continue to expand and increase in perimeter later during development confirmed the observations of some other research, but the exact nature of the governing forces of this phenomenon could not be fully explained by our data. It is likely to involve factors that have an impact on the rate that cellulose is laid down in the secondary cell walls of fibers, causing the fibers to expand throughout their development even though they are contained by the thin outer primary cell wall that is likely to be more flexible than the secondary cell wall. More controlled research addressing these issues would be of value and would certainly have implications on how decision support models are developed to allow fiber fineness properties like micronaire to be predicted.

Footnotes

Acknowledgements

Thanks go to Dr Greg Constable for supplying the genotypes used in this study, and to Jane Caton and Darin Hodgson for their technical assistance with the field experiments. We also thank Holly King, Glenda Howarth, Isabelle Miller, and Mark Freijah for their technical assistance regarding fiber quality testing and the manufacturing and testing of ring spun yarns. We acknowledge and thank Dr Bruce McCorkell for his assistance with the statistical analysis of data, and thank Dr Danny Llewellyn and Dr Stuart Gordon for their helpful comments and suggestions made during the preparation of this manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Partial financial support for this research was provided by the Cotton Research and Development Corporation through the project: CSP1308 “Agronomic Management for Better Fibre and Textile Quality”.

ORCID iD

Robert L Long

References

Wakelyn

Bertoniere

French

, et al. Cotton fiber chemistry and technology. Boca Raton: CRC Press, 2007.

Fowler

Hertel

KL.

Flow of a gas through porous media. J Appl Phys 1940; 11: 496–502.

Smith

WS.

Air gauge measures fiber fineness. Textile Industries 1947; 111: 86–88.

Lord

Air flow through plugs of textile fibers. Part II. The micronaire test for cotton. J Text Inst 1956; 47: T16–T47.

Montalvo

JG.

Relationships between micronaire, fineness, and maturity. Part I – Fundamentals. J Cotton Sci 2005; 9: 81–88.

Hequet

Wyatt

Abidi

, et al. Creation of a set of reference material for cotton fiber maturity measurements. Text Res J 2006; 76: 576–586.

Bange

van der Sluijs

MHJ

Constable

, et al. Fibrepak from seeds to good shirts – a guide to improving Australian cotton fibre quality. 2nd ed. Narrabri: The Cotton Research and Development Corporation, 2017, p.38.

Rodgers

Delhom

Fortier

, et al. Rapid measurement of cotton fiber maturity and fineness by image analysis microscopy using the Cottonscope®. Text Res J 2012; 82: 259–271.

Chapman

Staten

Cotton fiber maturity rapidly predicted with variable volume of sample in micronaire. Text Res J 1957; 27: 991–992.

10.

American Society for Testing and Materials, ASTM . Standard test methods for linear density and maturity index of cotton fibers (IIC-Shirley Fineness/Maturity Tester) D3818-92. In: Annual book of ASTM standards. Section seven: textiles. Vol 07.02. West Conshohocken: American Society for Testing and Materials, 1997, pp.133–136.

11.

Williams

Yankey

JM.

New developments in single fiber fineness and maturity measurements. In: Proceedings Beltwide cotton conferences (eds P Dugger and D Richter), Nashville, USA, 9–12 Jan 1996, pp.1284–1289. Memphis: National Cotton Council.

12.

Bradow

Wartelle

Bauer

, et al . Small-sample cotton fiber quality quantitation. J Cotton Sci 1997; 1: 48–60.

13.

Adedoyin

Toews

MD.

Characterization of single cotton fibers using a laser diffraction system. Text Res J 2010; 81: 355–367.

14.

Zhang

A novel method to assess cotton fiber qualities based on fraunhofer diffraction. In: Proceedings 2nd international Asia conference on informatics in control automation and robotics vol 1 (eds Y Wu and J Li), Wuhan, China, 6–7 March 2010, pp.226–228. Piscataway, NJ: IEEE Press.

15.

Seagull

Oliveri

Murphy

, et al. Cotton fiber growth and development 2. Changes in cell diameter and wall birefringence. J Cotton Sci 2000; 4: 97–104.

16.

Bange

Long

RL.

Impact of harvest aid timing and machine spindle harvesting on neps in upland cotton. Text Res J 2013; 83: 651–658.

17.

Bange

Long

Constable

, et al. Minimizing immature fiber and neps in upland cotton. Agron J 2010; 102: 781–789.

18.

Snipes

Baskin

CC.

Influence of early defoliation on cotton yield, seed quality, and fiber properties. Field Crop Res 1994; 37: 137–143.

19.

Bange

Constable

Johnston

, et al . A method to estimate the effects of temperature on cotton micronaire. J Cotton Sci 2010; 14: 164–172.

20.

Lokhande

Reddy

KR.

Quantifying temperature effects on cotton reproductive efficiency and fiber quality. Agron J 2014; 106: 1275–1282.

21.

Petkar

Mhadgut

Oka

PG.

Distribution of cross-sectional area, perimeter, circularity, and secondary wall thickness of different cotton fibers during growth. Text Res J 1986; 56: 642–645.

22.

Fiori

Brown

JJ.

Effects of cotton fiber fineness on the physical properties of single yarns. Text Res J 1951; 21: 750–757.

23.

Fiori

Brown

Sands

JE.

Effect of cotton fiber strength on single yarn properties and on processing behavior. Text Res J 1954; 24: 503–507.

24.

Simpson

Murray

MF.

Effect of cotton fiber fineness and strength on mechanical processing and open-end spinning and yarn properties. Text Res J 1978; 48: 270–276.

25.

Hunter

Predicting cotton yarn properties from fibre properties in practice. In: Proceedings 27th international cotton conference, (ed xxxx), Bremen, Germany, 24–27 March 2004, pp.62–77. Bremen: Faserinstitut.

26.

Hearn

Fitt

GP.

Cotton cropping systems. In: Pearson CJ (ed) Ecosystems of the world – field crop ecosystems. London: Elsevier, 1992, pp.85–142.

27.

Cotton Catchment Communities CRC. Australian cotton production manual. Narrabri: Cotton CRC, 2011.

28.

Constable

Shaw

AJ.

Temperature requirements for cotton. Agfact P5.3.5. NSW: Division of Plant Industries, New South Wales Department of Agriculture, 1988.

29.

Schleth

Ghorashi

Furter

Uster HVI 1000 application report – the role of cotton classification in the textile industry. Uster: Uster Technologies AG, 2007.

30.

United States Department of Agriculture, USDA Agricultural Marketing Services Cotton Program. Cotton classification understanding the data. Washington DC: United States Department of Agriculture, 2005.

31.

El Sourady

Worley

Stith

LS.

The relative contribution of fiber properties to variations in yarn strength in upland cotton, Gossypium hirsutum L. Text Res J 1974; 44: 301–306.

32.

Ureyen

Kadoglu

Regressional estimation of ring cotton yarn properties from HVI fiber properties. Text Res J 2006; 76: 360–366.

33.

Skau

EL.

Simple expressions for the circularity and fullness of fibers. Text Res J 1951; 21: 14–17.

34.

Hebert

Boylston

Wadsworth

JI.

Cross-sectional parameters of cotton fibers. Text Res J 1979; 49: 540–542.

35.

Lord

The characteristics of raw cotton. Manchester: The Textile Institute, 1961.

36.

Abbott

Higgerson

Long

, et al. An instrument for determining the average fiber linear density (fineness) of cotton lint samples. Text Res J 2010; 80: 822–833.

37.

Lynch

Michie

NA.

An instrument for the rapid automatic measurement of fiber fineness distribution. Text Res J 1976; 46: 653–660.

38.

Lunney

HWM

Irvine

PA.

Some factors affecting measurement by the CSIRO fiber-fineness distribution analyser. Text Res J 1979; 49: 371–379.

39.

Charlton

Sirolan-Laserscan. A review of its development, performance and application. Wool Tech Sheep Bree 1995; 43: 212–228.

40.

ASTM International. Standard test methods for diameter of wool and other animal fibers by Sirolan-Laserscan fiber diameter analyser D6466-10. In: Annual book of ASTM standards. Section seven: textiles. Vol 07.02. West Conshohocken: ASTM International, 2010, pp.590–598.

41.

Matic-Leigh

Cauthen

DA.

Determining cotton fiber maturity by image analysis: part 1: direct measurement of cotton fiber characteristics. Text Res J 1994; 64: 534–544.

42.

Thibodeaux

Rajasekaran

Development of new reference standards for cotton fiber maturity. J Cotton Sci 1999; 3: 188–193.

43.

Peirce

Lord

The fineness and maturity of cotton. J Text Inst 1939; 30: T173–T210.

44.

Kim

Rodgers

Delhom

, et al. Comparisons of methods measuring fiber maturity and fineness of upland cotton fibers containing different degrees of fiber cell wall development. Text Res J 2014; 84: 1622–1633.

45.

Cotton Incorporated. The classification of cotton. Cary: Cotton Incorporated, 2013.

46.

Balls

WL.

Studies of quality in cotton. London: Macmillan and Co., 1928.

47.

Faulkner

Hequet

Wanjura

, et al. Relationships of cotton fiber properties to ring-spun yarn quality on selected High Plains cottons. Text Res J 2012; 82: 400–414.

48.

Faerber

The spinning value of Australian cotton. Aust Cottongrower 2000; 21: 1–4.

49.

Ramey

Lawson

Worley

Relationship of cotton fiber properties to yarn tenacity. Text Res J 1977; 47: 685–691.

50.

May

Taylor

RA.

Breeding cottons with higher yarn tenacity. Text Res J 1998; 68: 302–307.

51.

Hake

Bragg

Mauney

, et al. Causes of high and low micronaire. Physiol Today 1990; 1: 1–4.

52.

Chauhan

Shah

Dweltz

NE.

A study of the distribution of perimeter, cell-wall area and thickness, and shape in raw and slack-mercerized single cotton fibers. Text Res J 1981; 51: 399–407.

53.

Hessler

LE.

The relationship between cotton fiber development and fiber properties. Text Res J 1961; 31: 38–43.

54.

Goynes

Ingber

Triplett

BA.

Cotton fiber secondary wall development – time versus thickness. Text Res J 1995; 65: 400–408.

55.

Pettigrew

WT.

Environmental effects on cotton fiber carbohydrate concentration and quality. Crop Sci 2001; 41: 1108–1113.

56.

Pettigrew

WT.

The effect of higher temperatures on cotton lint yield production and fiber quality. Crop Sci 2008; 48: 278–285.

57.

Xie

Trolinder

Haigler

CH.

Cool temperature effects on cotton fiber initiation and elongation clarified using in-vitro cultures. Crop Sci 1993; 33: 1258–1264.