This paper examines the nature of ‘reasonable’ argumentation in Generation of Animals III.10. Its aim is to develop an alternative to the dialectical construal of reasonable argumentation in Aristotle recently favoured by Robert Bolton. On the basis of a close textual analysis I show that the reasonable arguments deployed in Generation of Animals III.10 do not appeal to endoxa or reputable beliefs per se. Instead, they rely upon general facts about animals established by empirical induction. This implies that, contra Bolton, (...) not all reasonable arguments in Aristotle are dialectical; some, in fact, are thoroughly empirical and scientific. I conclude by suggesting that such empirical reasonable arguments even have a place in the De Caelo, the treatise that Bolton primarily focused upon when developing his interpretation. (shrink)