ABSTRACT.- Eucalyptus is one of the mainstays of the forest industry, contributing high-quality raw materials for pulp, paper, wood, and energy production. The typical approaches to reveal the genetic basis of important traits include classical Quantitative Trait Locus (QTL) mapping and Genome-Wide Association Studies (GWAS) approaches, but these are typically expensive and time-consuming. Here we evaluate the potential of Extreme-Phenotype GWAS (XP-GWAS) to identify candidate genes underlying a quantitative trait in Eucalyptus, using the timing of leaf heteroblasty as a case study. XP-GWAS involves genotyping pools of individuals grouped by extreme and opposed phenotypes from a population or a diversity panel and studying their allele frequency. Using a previous phenotyped trial of E. globulus, we sequenced pools of 50 individuals that notably differ in the onset of adult foliage. Since the genetic basis of heteroblasty is well understood, we first searched for previously identified genes. Secondly, we searched for new candidate genes and also evaluated the copy number variation (CNVs) that may be involved in this process. We found marginally significant SNPs associated with previously described microRNAs, and interesting new non-coding RNAs. Disease resistance genes were also uncovered, probably as a consequence of indirectly selecting resistant trees, although a possible interaction between resistance and heteroblasty cannot be disregarded either. Our work shows the utility and limitations of XP-GWAS analysis to explore the genetic basis of Eucalyptus. © 2023 Elsevier B.V.