Our understanding of maternal control of development in vertebrates remains incomplete. In this study, we investigated levels of maternal transcripts in good and poor quality eggs from artificially matured Japanese eel, using RNA‐Seq and quantitative polymerase chain reaction (qPCR), to identify candidate maternal transcripts related to development. De novo assembly or mapping of reads to the eel draft genome yielded 619,029 contigs and 85,906 transcripts, respectively; normalized read counts to these assemblies were calculated using reads (RPKM) or fragments (FPKM) per kilobase of transcript per million mapped reads. In silico screening identified 1,594 contigs and 150 transcripts with lower RPKM or FPKM in poor than in good quality eggs, 245 contigs, and 85 transcripts of which could be annotated by BLASTx, respectively. From selected contigs or transcripts, six genes (dnajb4, gnpat, card14, pdp1, fcgbp, ttn) had significantly lower messenger RNA levels in poor than in good quality eggs by qPCR. Multiple regression analysis showed that five genes (gnpat, b4galnt1, acsl6, rtkn, trim24) significantly correlated with hatchability. Taken together, 10 genes were identified as candidate maternal transcripts, regulating development in Japanese eel. Our results contribute to understanding the molecular basis for maternal control of development in vertebrates.