[2] The mariner's compass was not discovered till the twelfth or thirteenth century.

[3] This statement rests on the interpretation of Professor Finn Magnusen, for which see "The Voyages of the Northmen to America," Prince Socely's ed., pp. 34, 126. Boston, 1877. The general description of the climate and the products of the soil are in harmony with this interpretation, but it has nevertheless been questioned. Other Icelandic writers differ from him, and make the latitude of the land-fall of Leif at 49° 55', instead of 41° 43' 10", as computed by Magnusen.

This later interpretation is by Professor Gustav Storm. Vide The Finding of Wineland the Good, by Arthur Middleton Reeves, pp. 181-185. London, 1890. These interpretations are wide apart. Both writers are represented to be able and thorough scholars. When doctors disagree, who shall decide? The sciolists will doubtless range themselves on different sides, and fight it out to the bitter end.

The truth is, the chronology of that period in its major and minor applications was exceedingly indefinite. The year when events occurred is settled, when settled at all, with great difficulty; and it is plain that the divisions of the day were loose and indefinite. At least, they could only be approximately determined. In the absence of clocks, watches, and chronometers, there could not be anything like scientific accuracy, and the attempt to apply scientific principles to Scandinavian chronology only renders confusion still more confused. The terms which they used to express the divisions of the day were all indefinite. One of them, for example, was hirdis rismál, which means the time when the herdsmen took their breakfast. This was sufficiently definite for the practical purposes of a simple, primitive people; but as the breakfast hour of a people is always more or less various, hirdis rismál probably covered a period from one to three hours, and therefore did not furnish the proper data for calculating latitude. Any meaning given by translators touching exact hours of the day must, therefore, be taken cum grano salis, or for only what it is worth.

[4] It has been conjectured by some writers that Columbus on a visit to Iceland learned something of the voyages of the Northmen to America, and was aided by this knowledge in his subsequent discoveries. There is no evidence whatever that such was the case. In writing a memoir of his father, Ferdinando Columbus found among his papers a memorandum in which Columbus states that, in February, 1477, he sailed a hundred leagues beyond Tile, that this island was as large as England, that the English from Bristol carried on a trade there, that the sea when he was there was not frozen over; and he speaks also of the high tides. In the same paragraph we are informed that the southern limit of this island is 63° from the equator, which identifies it with Iceland. Beyond these facts, the memorandum contains no information. There is no evidence that Columbus was at any time in communication with the natives of Iceland on any subject whatever. There is no probability that he sought, or obtained, any information of the voyages of the Northmen to this continent. Ferdinando Columbus's Life of his father may be found in Spanish in Barcia's Historical Collections, Vol. I. Madrid, 1749. It is a translation from the Italian, printed in Venice in 1571. An English translation appears in Churchill's Collections, in Kerr's, and in Pinkerton's, but its mistranslations and errors render it wholly untrustworthy.

[5] It is somewhat remarkable that most writers who have attempted to estimate the value of the sagas as historical evidence have ignored the fact, that from a hundred and fifty to three hundred years they existed only in oral tradition, handed down from one generation to another, subject to the changes which are inevitable in oral statements. They are treated by these critics as they would treat scientific documents, a coast or geodetic survey, or an admiralty report, in which lines and distances are determined by the most accurate instruments, and measurements and records are made simultaneously. It is obvious that their premises must be defective, and consequently their deductions are sure to be erroneous.