I 4 4 IN STARRY REALMS. 



Viewed as a world, there is no more important point 

 for consideration than the length of the day as it would 

 appear to an inhabitant of Venus. This has long been a 

 matter of much uncertainty. The only means we have of 

 determining the period of rotation of a distant globe on 

 its axis is by watching some mark or object on the globe, 

 which we can recognise with certainty, and then observ- 

 ing the time which it requires to go round to the oppo- 

 site side, and back to where it was originally situated. 



On Mars this process presents but little difficulty. There 

 are many features on that globe which possess the desired 

 attributes of definiteiiess and sharpness ; and accordingly 

 the period of rotation of Mars on its axis, or the day of 

 that planet, is known with all desirable accuracy. It is 

 known even to the fraction of a second, and it happens by 

 a curious coincidence not to be very different from the 

 length of our own day. A similar method can be applied 

 to some of the other celestial bodies, though there are 

 individual peculiarities which often give rise to difficulties. 

 For instance, we know that the sun turns on its axis in 

 about twenty-five of our days. This is learned by the 

 observations of those spots by which the surface is fre- 

 quently marked. As, however, the sun is not a solid 

 body, and as the spots are merely apertures through a 

 covering of luminous clouds, they do not possess either 

 the defmiteness or the permanence which would enable us 

 to rely on them for any very minute accuracy. Each spot 

 is also too transient to enable any large number of conse- 

 cutive rotations to be watched by its aid ; and unless a 

 large number of rotations can be included in the observa- 

 tions as in the case of Mars, any very great precision in 



