168 MATHEMATICAL GEOGRAPHY. [CHAP.
the inference that the surface of the sea must be convex—a con-
clusion which would naturally involve the gradual revelation of
objects approaching upon it1. He does not however notice this;
still less does he use the fact as an argument to support the view
that the earth is circular. Strabo is the first author
who so employs it, and by him it is clearly stated.
of itf "It is evident," he says, "that, when persons on ship-
board are unable to see at a distance lights which are on a level
with the eye, the cause of this is the curvature of the sea; for, if
those lights are raised to a higher level, they become visible, even
though the distance be increased; and in like manner, if the
beholder attains a greater elevation, he sees what was previously
hidden...... Again, when men are approaching the land from the
sea, the parts near the shore-line come more and more in view,
and objects which at first appeared low attain a greater eleva-
tion2." Thus the evidences which were adduced to prove the
spherical form of the earth varied somewhat at different periods,
but the doctrine itself was accepted without question from the
time of Aristotle onward, and even before his age, by all scientific
men. The same thing is true of other points, in which the
principles of scientific geography depended more or less on astro-
nomical observations, such as the division of the globe into two
hemispheres by the equator, and the position of the lesser circles,
called the tropics, which are parallel to it. In every case, however,
the earth was regarded as forming the centre of the universe.
The idea of measuring the circumference of the earth seems
to have presented itself to the minds of the Greek
Philosophers at a comparatively early time. When
Aristophanes, in the Clouds, represents the disciple
of Socrates in his thinking-shop as saying that the object of
geometry was the measurement of .the whole earth8, he implies
1 Archimed., De iis quae in humido vehuntur, lib. i, probl. a. ; HOLVTQS
uSaros TJffvx&frrros war* axtvxjTov fUrw 17 £iri0ap«a cr^aipott&Jj gffrcu,
TO auro TJ? 70 xbrpov (Opera, vol. a, p. 357, ed. Heibeig).
* Strabo, 1.1. *o.
8 Aristoph., Nub. 202—4 :
MAS. yeuperpla. ZT. TOUT OUP
MA6. *w> ajKtfUTpeureau,. ST.
MAG. OUK, oXXd TTJV <rup,Tra<rcu>.