It may be said with general truth that the world-maps current up to the end of the 13th century had more analogy to the mythical cosmography of the Hindus than to any thing properly geographical.  Both, no doubt, were originally based in the main on real features.  In the Hindu cosmography these genuine features are symmetrised as in a kaleidoscope; in the European cartography they are squeezed together in a manner that one can only compare to a pig in brawn.  Here and there some feature strangely compressed and distorted is just recognisable.  A splendid example of this kind of map is that famous one at Hereford, executed about A.D. 1275, of which a facsimile has lately been published, accompanied by a highly meritorious illustrative Essay.[9]

82.  Among the Arabs many able men, from the early days of Islam, took an interest in Geography, and devoted labour to geographical compilations, in which they often made use of their own observations, of the itineraries of travellers, and of other fresh knowledge.  But somehow or other their maps were always far behind their books.  Though they appear to have had an early translation of Ptolemy, and elaborate Tables of Latitudes and Longitudes form a prominent feature in many of their geographical treatises, there appears to be no Arabic map in existence, laid down with meridians and parallels; whilst all of their best known maps are on the old system of the circular disk.  This apparent incapacity for map-making appears to have acted as a heavy drag and bar upon progress in Geography among the Arabs, notwithstanding its early promise among them, and in spite of the application to its furtherance of the great intellects of some (such as Abu Rihan al-Biruni), and of the indefatigable spirit of travel and omnivorous curiosity of others (such as Mas’udi).

[Sidenote:  Marino Sanudo the Elder.]

83.  Some distinct trace of acquaintance with the Arabian Geography is to be found in the World-Map of Marino Sanudo the Elder, constructed between 1300 and 1320; and this may be regarded as an exceptionally favourable specimen of the cosmography in vogue, for the author was a diligent investigator and compiler, who evidently took a considerable interest in geographical questions, and had a strong enjoyment and appreciation of a map.[10] Nor is the map in question without some result of these characteristics.  His representation of Europe, Northern Africa, Syria, Asia Minor, Arabia and its two gulfs, is a fair approximation to general facts; his collected knowledge has enabled him to locate, with more or less of general truth, Georgia, the Iron Gates, Cathay, the Plain of Moghan, Euphrates and Tigris, Persia, Bagdad, Kais, Aden (though on the wrong side of the Red Sea), Abyssinia (Habesh), Zangibar (Zinz), Jidda (Zede), etc.  But after all the traditional forms are too strong for him.  Jerusalem is still the centre of the disk of the habitable earth, so that the distance is as great from Syria to Gades in the extreme West, as from Syria to the India Interior of Prester John which terminates the extreme East.  And Africa beyond the Arabian Gulf is carried, according to the Arabian modification of Ptolemy’s misconception, far to the eastward until it almost meets the prominent shores of India.