30 DISTANCE AT WHICH MOUNTAINS 
does not depend solely upon the state of the low 
regions of the air, to which our meteorological ob- 
servations are confined, but also upon its transpa- 
rency and physical constitution in the most elevated 
parts; for the image is more distinctly detached, 
the more intense the aerial light which comes from 
the limits of the atmosphere has originally been, or 
the less it has lost in its passage. This in a certain 
degree accounts for the circumstance, that the Peak 
is sometimes visible and sometimes invisible to 
navigators who are equally distant from it, when 
the state of the thermometer and hygrometer is pre- 
cisely the same in the lower stratum of air. It is 
even probable, that the chance of perceiving this 
voleano would not be greater, were the cone equal, 
as in Vesuvius, to a fourth part of the whole height. 
The ashes spread upon its surface do not reflect so 
much light as the snow with which the summits of 
the Andes are covered, but, on the contrary, make 
the mountain, when seen from a great distance, 
become more obscurely detached, and assume a 
brown tint. They contribute, as it were, to equa- 
lise the portions of aerial light, the variable differ- 
ence of which renders the object more or less dis- 
tinctly visible. Bare calcareous mountains, sum- 
mits covered with granitic sand, and the elevated 
savannahs of the Andes, which are of a bright yel- 
low colour, are more clearly seen at small distances 
than objects that are perceived only in a negative 
manner ; but theory points out a limit beyond which 
the latter are more distinctly detached from the 
azure vault of the sky. 
The aerial light projected on the tops of hills in- 
creases the visibility of those which are seen posi- 
