In many animal species, elevated cAMP-PKA signaling initiates oocyte meiotic maturation upon hormonal stimulation, whereas in vertebrates, it acts as a negative regulator of this process. To address this “cAMP paradox”, we have focused on ARPP19 proteins. Dephosphorylation of Xenopus ARPP19 on a specific PKA site has been identified as a key step in initiating oocyte maturation. We first tracked evolution of the ARPP19 PKA phosphorylation site, revealing that it appeared early during the emergence of metazoans. This contrasts with strong conservation across eukaryotes of a phosphorylation site for the kinase Gwl in ARPP19 proteins, able to transform them into potent PP2A-B55 inhibitors and thus promote M-phase entry. We then compared the phosphorylation and function of Xenopus ARPP19 with its orthologue from the jellyfish Clytia, a model species showing cAMP-induced oocyte maturation. We confirmed that Clytia ARPP19 is phosphorylated on the conserved Gwl site in vitro as well as in maturing Xenopus and Clytia oocytes, behaving as a PP2A inhibitor and contributing to Cdk1 activation. However, Gwl-phosphorylated ARPP19 was unable to initiate oocyte maturation in Clytia, suggesting the presence of additional locks released by hormonal16 stimulation. Clytia ARPP19 was in vitro phosphorylated by PKA uniquely on the predicted site,17 but it was a much poorer substrate of PKA and of its antagonizing phosphatase, PP2A-B55delta, than the Xenopus protein. Correspondingly, PKA-phosphomimetic Clytia ARPP19 had a much weaker inhibitory activity on meiosis resumption in Xenopus oocytes than its Xenopus counterpart. Hence, poor recognition of Clytia ARPP19 by PKA and the absence of its targets in Clytia oocytes account for the cAMP paradox. This cross-species study of ARPP19 illustrates how initiation of oocyte maturation has complexified during animal evolution, and provides further insight into its biochemical regulation.