High-resolution spatial phenotyping of fibre diameter and staple length over Corriedale sheep fleeces

Abstract

The variability of raw wool traits over fleeces is still poorly known. We conducted a 2-year field study on 10 Corriedale ewes aimed at characterizing the spatial distribution of fibre diameter (FD) and staple length (SL) over fleeces. After ∼340 d of growth in two consecutive years we determined FD (μm; OFDA 2000) and SL (cm; ruler) on 128 staples (∼1.5 cm basal diameter) per animal collected over a regularly spaced grid. FD and SL varied between animals (FD: P < 0.001; SL: P < 0.05) and years (FD: P < 0.006; SL: P < 0.001). Variation between staples approached (FD, 80%) or surpassed (SL, 146%) variation between animals. Quadratic (FD, longitudinal) and quartic trends (SL, longitudinal; FD and SL transverse) best described the systematic variation along the axes of the fleece. Repeatabilities between years, from data pooled up to the animal × year level, were 0.88 ± 0.08 (estimate ± s.e., P < 0.001) for FD and 0.52 ± 0.24 (P < 0.05) for SL. Repeatabilities at individual sampling points over the fleece varied between 0.47 ± 0.26 and 0.94 ± 0.04 for FD, and from 0.0 ± 0.33 to 0.84 ± 0.10 for SL. Spatial ranges (distances required to consider two samples uncorrelated; estimated by variogram analyses) varied among animals (P < 0.04). They were higher (P < 0.03) for FD than for SL (5.2 vs. 3.7, s.e.d. = 0.46) and repeatable between years for both variables (FD: 0.87 ± 0.08, P < 0.001; SL: 0.72 ± 0.15, P < 0.01). Maps of traits produced by ordinary kriging exposed non-linear longitudinal and transverse trends of FD and SL over the fleeces, revealed a limited number of spatial features shared by most animals, and showed that some features of the spatial patterns uncovered are conserved over time. Spatial patterns and properties differed between FD and SL possibly reflecting both prenatally determined follicle populations and regionalized effects of environmental variables preferentially affecting SL. Maps of variance components and repeatabilities were bilaterally asymmetric suggesting the occurrence of waves of wool growth cycling over the fleece. Findings could be relevant for the design of experiments, fleece classing, selection of sampling points, and breeding. Alternative selection criteria and phenotyping methods would have to be developed for progressing from a fleece-oriented approach based on a single universal sampling point towards methods embracing 2D variability. Associated costs would limit applications to specialty wools.