his name to the monsoon. For information about India itself, the Greeks were, for a long time, dependent upon the account of Megasthenes, an ambassador sent by Seleucus, one of Alexander's generals, to the Indian king of the Punjab.

While knowledge was thus gained of the East, additional information was obtained about the north of Europe by the travels of one PYTHEAS, a native of Marseilles, who flourished about the time of Alexander the Great (333 B.C.), and he is especially interesting to us as having been the first civilised person who can be identified as having visited Britain. He seems to have coasted along the Bay of Biscay, to have spent some time in England,—which he reckoned as 40,000 stadia (4000 miles) in circumference,—and he appears also to have coasted along Belgium and Holland, as far as the mouth of the Elbe. Pytheas is, however, chiefly known in the history of geography as having referred to the island of Thule, which he described as the most northerly point of the inhabited earth, beyond which the sea became thickened, and of a jelly-like consistency. He does not profess to have visited Thule, and his account probably refers to the existence of drift ice near the Shetlands.

All this new information was gathered together, and made accessible to the Greek reading world, by ERATOSTHENES, librarian of Alexandria (240-196 B.C.), who was practically the founder of scientific geography. He was the first to attempt any accurate measurement of the size of the earth, and of its inhabited portion. By his time the scientific men of Greece had become quite aware of the fact that the earth was a globe, though they considered that it was fixed in space at the centre of the universe. Guesses had even been made at the size of this globe, Aristotle fixing its circumference at 400,000 stadia (or 40,000 miles), but Eratosthenes attempted a more accurate measurement. He compared the length of the shadow thrown by the sun at Alexandria and at Syene, near the first cataract of the Nile, which he assumed to be on the same meridian of longitude, and to be at about 5000 stadia (500 miles) distance. From the difference in the length of the shadows he deduced that this distance represented one-fiftieth of the circumference of the earth, which would accordingly be about 250,000 stadia, or 25,000 geographical miles. As the actual circumference is 24,899 English miles, this was a very near approximation, considering the rough means Eratosthenes had at his disposal.

Having thus estimated the size of the earth, Eratosthenes then went on to determine the size of that portion which the ancients considered to be habitable. North and south of the lands known to him, Eratosthenes and all the ancients considered to be either too cold or too hot to be habitable; this portion he reckoned to extend to 38,000 stadia, or 3800 miles. In reckoning the extent of the habitable portion from east to west, Eratosthenes came to the conclusion that from the Straits of Gibraltar to the east of India was about 80,000 stadia, or, roughly speaking, one-third of the earth's surface. The remaining two-thirds were supposed to be covered by the ocean, and Eratosthenes prophetically remarked that "if it were not that the vast extent of the Atlantic Sea rendered it impossible, one might almost sail from the coast of Spain to that of India along the same parallel." Sixteen hundred years later, as we shall see, Columbus tried to carry out this idea. Eratosthenes based his calculations on two fundamental lines, corresponding in a way to our equator and meridian of Greenwich: the first stretched, according to him, from Cape St. Vincent, through the Straits of Messina and the island of Rhodes, to Issus (Gulf of Iskanderun); for his starting-line in reckoning north and south he used a meridian passing through the First Cataract, Alexandria, Rhodes, and Byzantium.

The next two hundred years after Eratosthenes' death was filled up by the spread of the Roman Empire, by the taking over by the Romans of the vast possessions previously held by Alexander and his successors and by the Carthaginians, and by their spread into Gaul, Britain, and Germany. Much of the increased knowledge thus obtained was summed up in the geographical work of STRABO, who wrote in Greek about 20 B.C. He introduced from the extra knowledge thus obtained many modifications of the system of Eratosthenes, but, on the whole, kept to his general conception of the world. He rejected, however, the existence of Thule, and thus made the world narrower; while he recognised the existence of Ierne, or Ireland; which he regarded as the most northerly part of the habitable world, lying, as he thought, north of Britain.

Between the time of Strabo and that of Ptolemy, who sums up all the knowledge of the ancients about the habitable earth, there was only one considerable addition to men's acquaintance with their neighbours, contained in a seaman's manual for the navigation of the Indian Ocean, known as the Periplus of the Erythræan Sea. This gave very full and tolerably accurate accounts of the coasts from Aden to the mouth of the Ganges, though it regarded Ceylon as much greater, and more to the south, than it really is; but it also contains an account of the more easterly parts of Asia, Indo-China, and China itself, "where the silk comes from." This had an important influence on the views of Ptolemy, as we shall see, and indirectly helped long afterwards to the discovery of America.

PTOLEMAEI ORBIS

It was left to PTOLEMY of Alexandria to sum up for the ancient world all the knowledge that had been accumulating from the time of Eratosthenes to his own day, which we may fix at about 150 A.D. He took all the information he could find in the writings of the preceding four hundred years, and reduced it all to one uniform scale; for it is to him that we owe the invention of the method and the names of latitude and longitude. Previous writers had been content to say that the distance between one point and another was so many stadia, but he reduced all this rough reckoning to so many degrees of latitude and longitude, from fixed lines as starting-points. But, unfortunately, all these reckonings were rough calculations, which are almost invariably beyond the truth; and Ptolemy, though the greatest of ancient astronomers, still further distorted his results by assuming that a degree was 500 stadia, or 50 geographical miles. Thus when he found in any of his authorities that the distance between one port and another was 500 stadia, he assumed, in the first place, that this was accurate, and, in the second, that the distance between the two places was equal to a degree of latitude or longitude, as the case might be. Accordingly he arrived at the result that the breadth of the habitable globe was, as he put it, twelve hours of longitude (corresponding to 180°)—nearly one-third as much again as the real dimensions from Spain to China. The consequence of this was that the distance from Spain to China westward was correspondingly diminished by sixty degrees (or nearly 4000 miles), and it was this error that ultimately encouraged Columbus to attempt his epoch-making voyage.

Ptolemy's errors of calculation would not have been so extensive but that he adopted a method of measurement which made them accumulative. If he had chosen Alexandria for the point of departure in measuring longitude, the errors he made when reckoning westward would have been counterbalanced by those