THOUGHTS ABOUT KRAKATOA. 329 



Fortunately self-recording barometric instruments are now 

 to be found all over the world. Almost all the instruments 

 show distinctly the first great wave from Krakatoa to its 

 antipodes in Central America, and the return wave from 

 the antipodes to Krakatoa. They also all show the second 

 great wave which sped from Krakatoa, as well as the 

 second great wave which returned from the antipodes. 

 Thus, the first four of the oscillations are depicted on up- 

 wards of forty of the barograms. The fifth and sixth 

 oscillations are also to be distinguished on several of the 

 curves, and even the seventh is certainly established at some 

 few places, of which Kew is one. Then the gradually 

 increasing faintness of the indications renders them un- 

 recognisable, from which we conclude that after seven 

 pulsations our atmosphere had sensibly regained its former 

 condition ere it was disturbed by Krakatoa. 



Among the instruments which have yielded valuable 

 information about the air- wave, we have, curiously enough, 

 to mention the register of the recording gasometer-indica- 

 tor at Batavia. This apparatus, designed and employed 

 for a widely different purpose, shows that extraordinary 

 fluctuations in the barometric pressure occurred at the 

 time when the great wave passed over the town. 



It is of particular interest, from a physical point of 

 view, to study the numerical facts with reference to the 

 speed at which this world-embracing wave was propagated. 

 We shall for this purpose select the records taken at 

 Greenwich. The phase of the wave found most conve- 

 nient for measurement was the depression following the 

 outbreak, and the moment at which this phase started 

 from Krakatoa was 3 hrs. 32 mins. P.M. on August 17, 



