with yield in the replicated data between 1600 and 1800 kg ha−1 while there were none at that yield level in CSX211 and CSX218.Although the correlations are highly significant (P < 0.01), more so in population CSX206, there was deviation around the regression, indicating there would be potential for high yielding lines in PR to be selected or omitted incorrectly. However, the highest two lines for two years of replicated yield in population CSX206 were also highest in PR.
As with SPS, PR fibre quality data had significant associations(P < 0.01) with two year mean replicated data (data not shown).Fig. 3 shows the highly significant (P < 0.01) association between PR YQI and two year replicated YQI, with all populations having a