After the attempt had been made to determine the circumference of the earth, as was done by Eratosthenes with more or less satisfactory results, the thought, very naturally, was suggested of making an artificial representation of the entire earth, so far as then understood, that is, of making a terrestrial globe. There is no intimation, however, in any early allusion to Eratosthenes that he was a globe maker, or that he knew anything about globe construction. We know that he thought of the earth as a sphere placed in the center of the universe, around which the celestial sphere revolves every twenty-four hours.[23] Strabo, at a much later date, in referring to the geographical ideas of Eratosthenes, censured him for his unnecessarily elaborate proofs of the earth’s spherical character, apparently thinking the fact one too well known to require demonstration. [Illustration: Fig. 5. Globe according to Crates.] It appears to have been the grammarian Crates of Mallos, a contemporary of Hipparchus, and a member of the Stoic School of Philosophers, who made the first attempt to construct a terrestrial globe (Fig. 5), and that he exhibited the same in Pergamum, not far from the year 150 B. C.[24] It seems to have been Crates’ idea that the earth’s surface, when represented on a sphere, should appear as divided into four island-like habitable regions. On the one hemisphere, which is formed by a meridional plane cutting the sphere, lies our own oecumene or habitable world, and that of the Antoecians in corresponding longitude and in opposite latitude; on the other hemisphere lies the oecumene of the Perioecians in our latitude and in opposite longitude, and that of the Antipodes in latitude and longitude opposite to us.[25] Through the formulation and expression of such a theory the idea of the existence of an antipodal people was put forth as a speculative problem, an idea frequently discussed in the middle ages, and settled only by the actual discovery of antipodal regions and antipodal peoples in the day of great transoceanic discoveries.[26] That Strabo, at a later date, had this Pergamenian example in mind when stating certain rules to be observed in the construction of globes seems probable, since he makes mention of Crates’ globe. Strabo alone among ancient writers, so far as we at present know, treats of terrestrial globes, practically such as we find in use at the present day. He thought that a globe to be serviceable should be of large size, and his reasoning can readily be understood, for what at that time was really known of the earth’s surface was small indeed in comparison with what was unknown. Should one not make use of a sphere of large dimensions, the habitable regions (Fig. 6), in comparison with the earth’s entire surface, would occupy but small space. What Strabo states in his geography is interesting and may here well be cited. “Whoever would represent the real earth,” he says, “as