The concept of a species is quite important to an Aristotelian metaphysics. All the members of an Aristotelian species have qualitatively exactly the same nature. But it is difficult to figure out exactly which individuals are conspecific in this metaphysical sense.
It would be mistaken to think that metaphysical species are always co-extensive with biological ones. Biological species are defined in terms of populations that exhibit gene interchange. Thus, some biologists think there are six giraffe species. Suppose this hypothesis is right—if it's not right of giraffes, there will be some related case where it will be right. In nature, these different putative species of giraffes do not appear to interbreed, though in captivity they apparently do. If we thought metaphysical species to be coextensive with biological ones, we would have to answer tough questions about the offspring of different giraffe species. Suppose that the father is of metaphysical species A and the mother of metaphysical species B. Then any reason to think the offspring to be a metaphysical conspecific of the father is a reason to think it to be a conspecific of the mother, and these reasons seem to cancel out. So it seems better to suppose the offspring to be a member of a third metaphysical species. But if so, then with six species of giraffes, we could get up to 18 or maybe even 36 offspring metaphysical species. And when those bred, the metaphysical species would seem to keep on multiplying. This seems excessive, and so it seems better to suppose that, whatever the biologists say, the different giraffes are all members of one metaphysical species.
But all this reasoning strikes me as seriously ad hoc. If different biological giraffe species are metaphysically one species, why stop there? Why not suppose that all the members of the biological family Giraffidae, including giraffes and okapis, are one biological species?
It is not clear whether this is a metaphysical or an epistemological question. It might be that if we were clear enough on what metaphysical species are, we would make some progress here. On the other hand, people not sympathetic to Aristotelian metaphysics will take these difficulties as significant arguments against Aristotelianism.
I suspect that the answer here has a lot to do with teleology and the notion of normalcy. The Aristotelian nature encodes or defines how an organism ought to be, what its normal arrangement of parts is, what behavior is normal to it. Perhaps on the epistemological side, one can make a move here rather like that which David Lewis makes in regard to laws of nature. The metaphysical species provide a level of classification that offers the best balance between simplicity of normative description and richness of normative implications. If the metaphysical species included both giraffes and okapis, then the common nature would have to encode many conditional normative claims such as: if the individual has such-and-such features, then it should have a short neck, but if the individual has such-and-such features, then it should have a long neck. On the other hand, if giraffes and okapis are separate species, then we get a classification system with greater simplicity—the giraffe nature encodes having a long neck with no complicated conditionals, and the okapi nature encodes having a short neck again with no complicated conditionals.
How, then, do we know male and female giraffes to be members of one species? Because just about all the normative properties of males and females are the same, and we get a simpler classification system that just includes conditionals like if the individual has such-and-such features, then it should have ovaries, but doesn't reduplicate other normative claims.
But what about species, like Osedax worms, where male and female individuals seem to be very different? Maybe we Aristotelians just have to say that the balance of simplicity and richness account is only epistemic, and in this case is trumped by an analogical argument from other metaphysical species in which males and females are more easily seen as members of the same metaphysical species.