Remove observations if string part already occurs in another observation in the same group

Question

I am working with a biological data set with species sampled at different stations and dates. Species names consist of genus and species. If the species is not known, then o