84 



COGGIN BROWN: THE BURMA EARTHQUAKES OF MAY L9X2. 



Under these circumstances no reliance, whatever can be placed 

 on the times quoted by the Postmaster in Maymvo or on any 

 others based on " gun " or railway time there. There remain the 

 times given by the Director and the Deputy Superintendent of 

 Telegraphs, and it need only be pointed out that they are in agree- 

 ment with similar data from Mandalay. 



Most of the station masters in or near the epicentral tract 

 sent the times at which the station clocks stopped, or approximate 

 <niesses, evidently to the nearest live minutes. As examples llsnm- 

 hsai, Nawnghkio, Gokteik. Pyaunggaung, Kyaukme, and Bawgyo 

 may be quoted. The time records from the Soul hern Shan States 

 and the Ruby Mines District are rejected tor the same reasons. 



As far as can be ascertained from the data at my disposal 



therefore, the earthquake was felt in Mandalay 

 Time of commence- _„ ^ cu. j J nc ±1 



ment of shock. at 8-55 A.M., Burma Standard lime, on the 



morning of May 23rd. 1912, in Maymyo a 

 few seconds earlier, and in the epicentral area a lew miles to the 

 east of this, slightly earlier still. The probable time of commence- 

 ment of the shock therefore as accurately as can be determined, 

 lies between say 8-54 and 8-55 B.S.T. 



Burma possesses one seismograph which' is of the Omori type 



and is installed in Rangoon College. Other 

 Instrumental Records. . . , -, -, j r ±\ 



instrumental records were derived trom the 



seismographs of the Milne type in the meteorological observatories 



of Calcutta (Alipore), Kodaikanal in the Palni Hills and Bombay 



(Colaba), and from the Omori-Ewing seismographs in the same 



stations at Bombay and Simla. 



According to Middlomiss : — 



" If recent oritioism of Beismographio records is to he trusted (us to whioh 

 specialists in this branch of science can alone speak with particular know- 

 ledge), the Milne Beismograph trace, which is very small and often hlurred, 

 cannot be trusted to show all those minute sul>-di visions of regularly recur- 

 ring period and amplitude which the larger forms working with a large 

 natural period of swing, register by means of a needle point on smoked 

 paper, and which give an open or large scale diagram. 



The SO-called preliminary tremors as seen in long-distance seismograms 

 written by the Milne instrument, are, however, fairly well differentiated from 

 the large movement which follows. The splitting of the preliminary tremors 

 into two groups, first and second, is also believed by many to be sufficiently 

 recognisable. However that may be. the beginning of the largo movement 

 is the only definite point that can be reasonably correlated with the sensible 



