This article interprets the idea in a very narrow fashion: that ontogeny must strictly follow phylogeny in every aspect, otherwise the idea is dispoved. But tha

This article interprets the idea in a very narrow fashion: that ontogeny must strictly follow phylogeny in every aspect, otherwise the idea is dispoved. But that strict version is obviously not true, of course evolution can add new features at an early stage of ontogeny, like the ecidna limbs mentioned in the article. I interpret the "ontogeny recapitulates phylogeny" just as the observation that some embryos look like fish, and then they look like amphibians. That is just an observation, and I think it is interesting, and I think it spurred a lot of research into how organisms develop. It seems many of the patterns that nature uses are conserved. And it is "easier" to add another feature on top of a certain foundation, than to change something early in development, which could topple all the following steps.