106 REPORT— 1900. 



University ; the observations made at Zikawei indicate that at certain 

 observatories at least the unfelt movements of earthquakes may from time 

 to time have serious effects on timekeepers. 



With the object of throwing light upon this subject I shall esteem 

 it a favour if directors of observatories will let me know whether any 

 changes were observed or not observed in the rates of pendulum time- 

 keepers on dates corresponding to those of large earthquakes enumerated 

 on p. 108, addressing their communications to me at Shide, Isle of Wight, 

 England. 



IV. Earthquakes and Rain. 



In the British Association Reports for 1899, p. 209, I gave a quota- 

 tion from Mr. O. H. Howarth respecting a heavy condensation of aqueous 

 vapour which he observed for three hours after the Mexican earthquake 

 of January 24, 1899. This was in the form of a heavy mist which settled 

 over the head of a canon at an elevation of 8,700 feet. 



Mr. Howarth states that in this place such mists are never seen at this 

 time of the year, it being the middle of the dry season. 



Something similar to this occurred on June 12, 1897, after the severe 

 earthquake which originated on that day in the highlands of Assam. 

 Mr. H. Luttman-Johnson, I.C.S., in the 'Journal' of the Society of Arts, 

 April 15, 1898, describes the weather before the earthquake as having 

 cleared : the afternoon was lovely, and there was not a cloud in the sky. 

 Five minutes after the earthquake the residents in Shillong were sur- 

 rounded with cloud and mist, and they sat up all night with rain beating 

 upon all sides. 



Captain A. A. Howell, I.C.S., deputy-governor of the Garo Hills, 

 gives the actual rainfall. "The records taken at 8 a.m. showed that for 

 the twenty-four hours preceding the 12th there was no rain. There was 

 rain at noon on the 12th, but it cleared off at 2 p.m. The earthquake 

 occurred at about 5 p.m., and after that until next morning 3'26 inches 

 fell. 



In considering whether there is any possibility of a connection 

 between the phenomena here considered we must remember that observa- 

 tions showing that rain and cloud have followed closely on the heels of 

 certain earthquakes appear to be confined to tropical and semi-tropical 

 countries ; and it is in these countries where sudden showers, indicating the 

 collapse of critical atmospheric conditions, are frequent. Given, therefore, 

 such conditions at no great distance above the surface of the earth, which 

 was probably the condition in the highlands of Assam, and then admit 

 that beneath the gaseous covering consisting of layers of air of different 

 temperatures and with different degrees of saturation 10,000 square miles 

 of mountainous country was moved, or that a much larger ai-ea was thrown 

 into violent wave-like movement, we recognise that the relationship of 

 earthquakes and atmospheric precipitation may not be so improbable as is 

 generally supposed. As the ground rose upwards, the air immediately 

 above it would suffer compression, and as the ground fell there would be 

 rarefaction, whilst layers of air differing in their physical state might be 

 mixed, and a vigorous seismic activity might in this way result in pre- 

 cipitation. 



