Populus ussuriensis Kom. is one of the most important tree species for forest renewal in the eastern mountainous areas of Northeast China due to its fast growth, high yield, and significant commercial and ecological value. The selection of optimal reference genes for the normalization of qRT-PCR data is essential for the analysis of relative gene expression. In this study, fourteen genes were selected and assessed for their expression stability during abiotic stress (drought, high salinity, and cold stress) and after the treatment with the drought-related hormone ABA. Three algorithms were used, geNorm, NormFinder, and BestKeeper, and a comprehensive ranking of candidate reference genes was produced based on their output. The most appropriate reference genes were UBQ10 and RPL24 for drought and ABA treatment, UBQ10 and TUB3 for cold stress, and UBQ10 and 60S rRNA for high salinity. Overall, UBQ10 was the most stable reference gene for use as an internal control, whereas PP2A was the least stable. The expression of two target genes (P5CS2 and GI) was used to further verify that the selected reference genes were suitable for gene expression normalization. This work comprehensively assesses the stability of reference genes in Populus ussuriensis and identifies suitable reference genes for normalization during qRT-PCR analysis.