2 Neioconib and Dutton — Speed of 



country have an agent or reporter at "Washington and he usually 

 keeps a file of the paper he serves. The library of Congress 

 keeps files of two or more papers from every State. As many 

 of these as practicable were thoroughly examined. Many local 

 papers were requested to furnish copies of their issues of Sept. 

 1st, 2d and 3d, and most of them complied. Many marked 

 copies of papers were sent to the survey from unexpected 

 sources. Altogether more than four hundred time reports 

 were gathered. 



As might be expected a portion of these were useless. In 

 order that it may be apparent which were selected for considera- 

 tion and which rejected, the following account is given. There 

 were about thirty which stated that the shock occurred " about 

 10 o'clock " or " a few minutes before 10." As a single min- 

 ute is a very important quantity here, all such reports were 

 summarily rejected as too indefinite. The reports from light- 

 houses in most cases proved unavailable. These structures 

 being situated most frequently where access to standard time 

 is difficult, their clocks are regulated by the sun and an 

 almanac. The uncertainties of this method of time keeping 

 were evidently too great to justify any attempt to utilize 

 them. But a few lighthouses keep standard time and in all 

 such cases their reports were admitted to consideration. There 

 were a few (nine or ten) which gave times so widely aberrant, 

 differing by a quarter to half an hour from the great mass of 

 records, that they were rejected. The whole number which 

 received preliminary consideration was 316, many of which it 

 was expected would also be rejected after a more thorough 

 examination, due cause being assigned. These 316 observa- 

 tions were catalogued in alphabetical order, the latitudes and 

 longitudes Of the localities being roughly ascertained and also 

 their distance from the centrum. 



By far the most important time determination is that of the 

 centrum, which was computed to be about six seconds earlier 

 than that of Charleston. The time at Charleston is derived as 

 follows. Among the numberless clocks stopped in that city 

 by the earthquake, there were four which had compensated 

 seconds pendulums and second hands and were of the pattern 

 generally classed as " jewelers' regulators." All were compared 

 daily with the time signal of the Western Union Telegraph Co., 

 and the testimony is positive that none of them had errors on 

 August 31st exceeding nine seconds, while the mean probable 

 error of the four was certainly much less than this. The first 

 was the regulator of James Allan & Co., Jewelers, No. 285 

 King street. It was regulated by means of a " sounder," which 

 was daily put into circuit with the Western Union time signal 

 wire. The clock was corrected only when its error exceeded 



