Industrial Crops and Products 102 (2017) 16–22 Contents lists available at ScienceDirect Industrial Crops and Products jo u r n al homep age: www.elsevier.com/locate/indcrop The impact of fiber maturity on estimating the number of cotton (Gossypium hirsutum L.) fibers per seed surface area Addissu Ayele a,∗ , Eric Hequet a , Brendan Kelly a,b a Fiber and Biopolymer Research Institute, Texas Tech University, Lubbock, TX 79409, USA b Texas A&M AgriLife Research & Extension Center at Lubbock, TX 79403, USA a r t i c l e i n f o Article history: Received 16 September 2016 Received in revised form 27 February 2017 Accepted 3 March 2017 Keywords: Cultivar Fiber density Fibers per seed Fiber maturity Fiber length distributions a b s t r a c t The number of fibers per seed surface area (fiber density) is a selection criterion used to improve cotton fiber quality and yield simultaneously in cotton breeding programs. However, the parameters utilized to estimate fiber density are calculated from fiber quality attributes that are sensitive to environmental variations, especially fiber maturity. Fiber maturity is one of the most important fiber properties that influences other fiber properties such as individual fiber strength and length. In order to investigate the impact of within-plant fiber maturity on the estimates of fiber density, field experiments were conducted at Lubbock, Texas during the 2012, 2013 and 2014 growing seasons. A set of twelve upland cotton cultivars popular in the Texas High Plains were grown in a randomized complete block design with three field replications. Bolls were box picked at harvest to provide samples from each cultivar representing a range of fiber maturity. Fuzzy seeds obtained after roller ginning were acid-delinted, scanned on a flatbed scanner, and the WinSeedle Pro software was used to estimate the seed surface area (SSA). The number of fibers per seed was estimated using the fiber quality parameters provided by AFIS (Advanced Fiber Information System), the lint weight, and the number of seeds in the sample. The number of fibers per seed surface area was obtained by dividing the number of fibers per seed by the estimated SSA. In this study, cultivars with relatively less stable fiber maturity across the fruiting branches of the cotton plant tend to have more variable number of fibers per seed surface area estimates, suggesting that the calculation of the number of fibers per seed surface area may be biased. It is likely that the source of bias is fiber maturity (immature fibers are weak and tend to break during mechanical processing). Therefore, fiber maturity should be considered when screening lines based on estimated fiber density because low fiber maturity can lead to unreliable estimates of fiber density. © 2017 Elsevier B.V. All rights reserved. 1. Introduction International spinning market requirements along with advances in spinning technologies have increased demand for cotton (Gossypium hirsutum L.) fibers that are long, uniform, mature, fine, strong, and with low contamination (such as neps, bark, and seedcoat fragments). Domestic producers must target these cotton fiber quality attributes to remain competitive in the currently dominant ring spinning market. However, gains in cotton fiber quality should not be made at the expense of yield. Breeders developing cultivars competitive on international markets should select for not only improved fiber quality but also improved yield. ∗ Corresponding author. E-mail addresses: addissu.ayele@ttu.edu, mahideraddissu@gmail.com (A. Ayele). Nevertheless, it is a great challenge for cotton breeders to improve lint yield and fiber quality simultaneously because of the negative association between lint yield and fiber quality. Smith and Coyle (1997) suggested that within-boll lint yield is based on the number of spinnable fibers produced within the boll but Upper Half Mean Length (UHML) was negatively associated with fibers per seed and fiber per seed surface area in the F1 populations. This neg- ative association suggests that selection for an increase in lint yield will result in a decrease in UHM. Zeng and Meredith (2009) reported that fiber length and strength have also been shown to be negatively associated with lint yield. There is considerable evidence for a neg- ative relationship between many important fiber properties and lint yield (Meredith and Bridge, 1971; Tyagi, 1987; Clement et al., 2012). Without the proper selection criteria, the process of devel- oping cultivars with improved lint yield may come at the expense of fiber quality. Designing better strategies to improve fiber quality http://dx.doi.org/10.1016/j.indcrop.2017.03.004 0926-6690/© 2017 Elsevier B.V. All rights reserved.