Hamilton. — On our Knoivledge of the Oceanic Areas. 173 



The strait pass where Hercules ordain'd 



The boundaries not to be overstepped by man ;* 



beyond which lay the "deep inimitable main," "the un- 

 peopled world," of which the learned as yet knew nothing. 

 Eastward the limit was fixed at the mouth of the Ganges. 

 In this direction, therefore, mediaeval geography, as it stood 

 towards the close of the thirteenth century, had not only not 

 advanced beyond the point at which Ptolemy left it, but had 

 actually receded. 



Although the usual form of the habitable world as depicted 

 in the Middle Ages was circular, a quadrangular shape was 

 sometimes adopted, based upon too literal an acceptation of 

 the passage of the Scripture which speaks of the " four corners 

 of the earth." There is yet another form in which a map 

 was constructed, and which was perhaps more correct. On 

 the Matthew Paris maps we are told that the world in its 

 truest form resembles an extended military cloak {chlamys 

 extensa). The chlamys consisted of a central square with 

 wings added to it, wider at the bottom than at the top, the 

 whole shape being a greatly truncated triangle. This idea 

 was probably derived from Macrobius,! who in his turn bor- 

 rowed it from Strabo (ii., p. 113). 



Another point of interest is the orientation of the maps. 

 Our predecessors, with few exceptions, placed the east in that 

 position at the top of the map. Biblical considerations again 

 decided this. The primeval abode of man was in the east, the 

 terrestrial Paradise still remained there. On this subject of 

 the location of the terrestrial Paradise there is a large mass of 

 mediaeval literature ; but in the whole of it there is no doubt 

 of its being an existing contemporaneous fact. Mandeville 

 (cap. XXX.) says that he had not visited it himself on ac- 

 count of his unworthiness, but he describes it at length on 

 the information of trustworthy persons. The four rivers of 

 Paradise were usually identified with the Euphrates, Nile, 

 Ganges, and Tigris, and the difficulty as to the widely remote 

 sources of these rivers was solved by assuming that the rivers 

 on leaving Paradise were submerged, and reappeared at these 

 points. 



The traces of this belief are to be seen even in the person 

 of Columbus, for we learn in Irving' s " Life of Columbus," book 

 iv., chapter 4, that when the great navigator encountered the 

 flood of the River Orinoco, in the Gulf of Paria, he thought it 

 could be none other than the fount of Paradise. 



Of the renaissance of enterprise and the desire for know- 



* Dante, " Inferno," xxvi. 



t De Somn. Scip., ii., 9, vvtere Macrobius is commenting ou Cicero's 

 description, "Augusta verticibiis, lateribus latior" (De Republica, vi., 20). 



