1 88 LIGHT SCIENCE FOR LEISURE HOURS. 



in a sea which so far as they were concerned might be 

 shallow, yet if it is known that a single object has 

 floated in it which was not flat, but on the contrary 

 had great length, and breadth, and thickness, we know 

 that the sea must be a deep one. Some among the 

 rounded clouds of Jupiter, which not only by their 

 shape, but by their shading, indicate a globular figure, 

 would, if actually globular, require an atmosphere 

 five or six thousand miles deep at the very least. The 

 atmosphere may not be so deep as that, or may be 

 very much deeper. Certainly it would at once remove 

 the difficulty last considered if we could suppose the 

 cloud-bearing atmosphere of Jupiter to be thirteen or 

 fourteen thousand miles in depth, for then the solid 

 globe within would not differ very much in mean 

 density from the globe of our earth. But supposing 

 we assume, as the result of the actual telescopic aspect 

 of the cloud-belts, the depth of the atmosphere to be 

 but about 2,000 miles, which would be less than the 

 apparently minute diameter of one of the satellites, we 

 should even then find that under the tremendous 

 pressure exerted by Jupiter's attraction the lower 

 strata of such an atmosphere, if composed of any gases 

 known to us, and at the temperature of our own air 

 even in the torrid zones, would be simply compressed 

 into the solid or liquid form. At least they could not 

 continue to obey the laws which perfect gases obey 

 under pressure. Assuming the pressure at the visible 

 limit of the cloud envelope to be less than one- 

 thousandth part of the pressure of our air at the sea- 



