The only suggestions which attempt to deal with the reading of the manuscripts are that of Clodachzh (‘in search of spices’, which in addition to its improbability uses propter in a causal, and not in a local sense; the latter alone appears to be the Lucretian usage) and that of Ellis, ‘which does preserve most of the letters but shuffles them up in an impossible manner’ (Bailey) and for this reason has not found acceptance. Common ground for emendation is the need to provide a substantive with the meaning ‘ships’ for velivolis. It follows that if this is to be done, and if at the same time anything resembling the manuscripts' propter odores is to be retained, then a lacuna containing navibus must be presumed. (There is no help to be obtained from ii. 417, which does no more than record the existence of propter odores, words which there fit the context.)