LIGHT. 121 



degree of rarefaction, the atmosphere below the observed 

 height being too dense and that above too rare to allow the 

 electricity to pass. 



The odour which many metals, such as iron, tin, and zinc, 

 emit, and the so-called thermographic radiations, can hardly 

 be explained upon any other theory than the evaporation of 

 an infinitesimally small portion of the metal itself. 



So universal is the tendency of matter to diffuse itself into 

 space, that it gave rise to the old saying that nature abhors a 

 vacuum ; an aphorism which, though cavilled at and ridiculed 

 by the self-sufficiency of some modern philosophers, contains 

 in a terse though somewhat metaphorical form of expression, 

 a comprehensive truth, and evinces a large extent of obser- 

 vation in those who, with few of the advantages which we 

 possess, first generalised by this sentence the facts of which 

 they had become cognisant. 



It has been argued that if matter were capable of infinite 

 divisibility, the earth's atmosphere would have no limit, and 

 that consequently portions of it would exist at points of space 

 where the attraction of the sun and planets would be greater 

 than that of the earth, and whence it would fly off to those 

 bodies and form atmospheres around them. This was sup- 

 posed to be negatived by the argument of the well-known 

 paper of Dr. Wollaston ; in which, from the absence of notable 

 refraction near the margin of the sun and of the planet 

 Jupiter, he considered himself entitled to conclude that the 

 expansion of the earth's atmosphere had a definite limit, and 

 was balanced at a certain point by gravitation : this deduction 

 has been shown to be inconclusive by Dr. Whewell, and has 

 also been impugned upon other grounds by Dr. Wilson. There 

 is a point not adverted to in these papers, and which Wol- 

 laston does not seem to have considered, viz. that there is no 

 evidence that the apparent discs of the sun and of Jupiter 

 show us their real discs or bodies. Sir W. Herschel regards 

 the margin of the visible discs as that of clouds or a peculiar 

 state of atmosphere, and the rapidly changing character of the 

 apparent surfaces renders some such conclusion necessary. If 

 this be so, refraction of an occulted star could not be detected 

 at all event c . ?V the denser portion of the atmosphere. . 



