57 



Toiiud tlie earth "botli east and west immediately after the 

 Krakatoa explosion. After reductions for temperature he 

 found that the excess of velocity of waves travelling east over 

 the velocity of waves travelling west was about 32 miles an hour, 

 and he states how this may be explained by the fact that the 

 winds along the wave paths would on the whole be from the 

 west, which would cause an increase in the rate of one set and 

 decrease in that of the other, so that the difference of 32 miles 

 would correspond to an average westerly wind of 16 miles au 

 hour. 



Now, the dates on which the phenomenon was recorded at 

 various places necessitate a rate of progression averaging 2,000 

 miles a day, and we can hardly imagine this average increase 

 being set up over at least half the equator so immediately 

 after the eruption, whilst from various persons we get the 

 news that these colours were seen before the date of the 

 eruption even commencing. 



xA.t the 3Iathematical Society, Port Adelaide, February 27th, 

 when the subject was before that Society, I stated that I had 

 seen similar appearances in July, 1S83, especially in the morn- 

 ings, and also that in February, when at Port Victoria, I wit- 

 nessed some of the most beautiful rosy sunsets, which I 

 likened in my letters home to the fairy transformation scenes. 



Our Astronomer Eoyal, Mr. Chas. Todd, C.M.G., stated that 

 he had news of these appearances occurring before the date of 

 the explosion. 



JMr. Xeison, Grovernment astronomer of Xatal, says that 

 iihey commenced there in February, 1883, but on a less grand 

 scale, gradually becoming more marked until June. For the 

 next two months nothing was noticed. At the end of August 

 they became most vivid — on the 2ist and 22nd noticeable but 

 not vivid, next five days stormy with much rain and lightning, 

 on 2Sth and 29th vivid, 30th rainy, 31st vivid. In September, 

 1st, 2nd, 3rd, and 5th vivid, and most remarkably fading away 

 into green and purple in the east. Then a week of much rain, 

 and the displays vanished for four months. 



Xow for a remarkable point. In the Transvaal they were 

 first noticed on September 2, and were prominent there whilst 

 invisible at Xatal, though the places were onl}^ separated by a 

 distance of 250 miles. 



S^^duey Hooper states that he has observed sunsets for 30 

 3^ears, and has always found the crimson glow coming a con- 

 ■siderable time after other tints. 



E. Ean storm, a landscape painter, states that from 30 years 

 experience he is certain similar displays take place ever}' year. 



The analysis of snows and rains has given entirely different 

 results in various cases : — 1. The analyses by Nordeuskoold, the 



