42 SCIENCE IN GRECO-ROMAN ANTIQUITY 



accumulates on one half of this sphere, making it 

 luminous ; the other half remains dark. This is why 

 the earth, placed in the centre of the universe, sees the 

 alternation of day and night. As to the sun, it is 

 merely the image of the earth, produced by reflection. 

 " The sun is not a fiery substance, but an image of 

 reflected flame, similar to that which comes from water ' 

 (Diels, Vor. I, p. 158, 35). The light which comes from 

 the fiery hemisphere strikes the earth, then, concen- 

 trated there, it is sent back on to this same hemisphere, 

 where it appears to us as a luminous disk. That this 

 really was the idea of Empedocles, Plutarch confirms 

 by one of the characters he introduces ; " You laugh 

 at Empedocles," said he, " because he attributes to the 

 sun the following origin : the light of the sky after 

 having been reflected on the surface of the earth, reflects 

 the image of the earth again on the sky " (Diels, Vor. 

 I, p. 188, 8). This conception, although at first sight 

 curious, is very easily explained. 1 The discovery had 

 just been made that the moon shone by reflected light, 

 and Empedocles was naturally led to give to this theory 

 a wider application than was legitimate. 2 



1 8 Burnet, Aurore, p. 272. 



2 It is interesting to compare the views of Empedocles with 

 the ideas expressed by the astronomer Nordmann in his 

 scientific romance, entitled " Einstein and the Universe." 

 The curvature of space being constant and such that it closes 

 upon itself like a spherical surface, one may imagine "that 

 the rays emanating from a star, from the sun, for example, 

 will converge at a diametrically opposite point of the Universe, 

 after having gone round it," and that they thus form a new 

 star. It is true, adds Nordmann, we have not yet been able 

 to prove the existence of these phantom stars, " But what 

 observers could not do yesterday, they will be able to do 

 to-morrow by the help of the suggestions of the new science." 

 One can thus foresee " the surprising and unexpected conse- 

 quences of the new conceptions, which exceed in their fan- 

 tastic poetry all the most romantic constructions of imagina- 

 tive extrapolation. The real or at least the possible ascends 



