THOUGHTS ABOUT KRAKATOA. 327 



globe has the stupendous dimensions expressed by a dia- 

 meter of 8,000 miles, and imagine it to be enclosed in a 

 uniform shell of air. Now, suppose that all is quiet, till 

 at some point, which for the moment we may speak of as 

 the pole, a mighty disturbance is originated. Let us 

 regard this disturbance as produced by a sudden but local 

 pushing up of the atmosphere by a force directed from the 

 earth's surface outwards, and let us trace the effect thereby 

 produced on the atmosphere. Such a sudden impulse will 

 at once initiate a series of circular atmospheric waves, 

 which will speed away from the centre of disturbance 

 just like the waves caused by the pebble in the pond. If 

 the original atmospheric impulse be large enough we shall 

 find the circle growing larger and larger, its radius in- 

 creasing from hundreds of miles to thousands of miles, 

 until at last the wave reaches the equator. What is to 

 happen when the diverging waves have attained the 

 equator, and are now confronted by the opposite hemi- 

 sphere ? This is one of those cases in which the mathe- 

 matician can guide us where the experimentalist would be 

 otherwise somewhat at fault. We know that as the wave 

 entered the opposite hemisphere it would at once move 

 through a similar series of changes to those through which 

 it had already gone, but in the inverse order. The wave 

 will thus, after leaving the equator, glide onwards into a 

 parallel small circle, ever decreasing in diameter, and con- 

 verging toward the anti-pole. Finally, just as the waves 

 all radiated from the original pole, so will they all con- 

 centrate towards the opposite one. But what is now to 

 happen ? Here, again, the mathematician will inform us. 

 He can follow the oscillations after their confluence, and 



