However this may be, the identification of the wild corn will always be an insoluble problem. The older commentators on these sagas used to consider that maize was indicated, but this is not, properly speaking, a wild plant, and moreover bears singularly little resemblance to any European cereal. The later school mostly identifies the corn of the sagas with wild rice, but this is open to the objection that it is an aquatic plant. On the whole, therefore, while I think the discovery of the vine is indisputable, and was the cause rather than the effect of any trace of the influence of the legends of the Insulae Fortunatae to be met with in the sagas, I confess, in spite of the coincidence of the reports of later explorers, to regarding the corn as a more difficult problem. In any case it seems to me that the absence of all mention of wild corn in the Flatey Book version has a most significant bearing on Dr. Nansen’s argument. For in practically all references to the Fortunate Islands the corn and the vines are so closely connected that a borrower from such sources could hardly take the one without the other. E.g. Horace, _Epodes_, xvi. 41: ‘Beata petamus arva, divites et insulas; reddit ubi _Cererem_ tellus inarata quotannis, et imputata floret usque _vinea_’; and Isidore, _Etymologiarum_ xiv. 6: ‘Fortuitis _vitibus_ iuga collium vestiuntur; ad herbarum vicem _messis_.’ The existence, therefore, of a circumstantial account of Wineland, which contains no mention of wild corn, makes any derivative connexion between the descriptions of this country and the Insulae Fortunatae, apart from all other difficulties, exceedingly improbable. _Celtic Legends._ When we turn to the other features of the saga, we find Dr. Nansen displaying even greater resource and ingenuity in finding parallels in the folk-lore of other lands. The argument from analogy is proverbially untrustworthy, but it is at the same time rather difficult to combat effectively where, as in the present case, it is impossible to set out the full number of alleged resemblances with which Dr. Nansen’s industry in research has provided him. Samples are open to the charge of unfair selection. I should doubt, for example, whether even Dr. Nansen himself, though he emphasizes the parallel with a marginal heading, can attach any real importance to such an instance as the following: