Paraphyletic and polyphyletic taxa in the hydroidomedusae result in part from the traditional assignment of similar species to separate genera according to presence or absence of the medusa stage. Medusa reduction is widespread in the Antho- and Leptomedusae and has occurred repeatedly in the same clade. Petersen raised this issue for the Anthomedusae-Capitata, with a thorough taxonomic re-arrangement of the whole group. Advancing knowledge of life cycles allows identification of poly- and paraphyly in other hydrozoan groups, so that phylogenetic interpretation of some "anomalies" in the classification of Antho- and Leptomedusae is possible. Most difficult cases are those leading to "inconsistent" evolution. It is here stressed that the "inconsistency" is due to different rates of change of the two main morphs of the cycle, whereas it is obvious that the species as a whole evolve consistently. The cases of supposed inconsistent evolution considered here are: Zancleidae vs. Corynidae (similar hydroids - different medusae); Anthomedusae vs. swimming gonophores (different hydroids - similar medusae); Clytia vs. Obelia (similar hydroids - different medusae); Stomotoca vs. Hydrichthys (similar hydroids - different medusae); Zancleidae vs. Teissieridae (different hydroids - similar medusae); Clytia vs. Phialella (different hydroids - similar medusae); Mitrocomium-Campalecium vs. Eucheilota-Lovenella (different hydroids - similar medusae); Eutimu-Eirene vs. Eugymnanthea (similar hydroids - different medusae). In some cases the resemblances are due to convergence, in other cases it is possible to recognize homologies. Resolution of paraphylies, however, may prove difficult. Laomedea, for instance, could represent paedomorphic species of both Cljtiu and Obelia species, since the hydroids are similar and the difference resides almost exclusively in the gonothecal content. Taxonomy must represent phylogenetic relationships among taxa, but has also to be a useful tool to name taxa and one of its included aims is nomenclatura1 stability. These two goals are difficult to achieve in hydroidomedusan systematics, especially because the rate of discovery of new taxa and of new life cycle types is still so high that the overall picture is far from being completed. For the sake of stability, it is suggested not to emend generic distinctions leading to supposed paraphyly until the features of the species of such genera have been fully studied.