September 8, 1905.] 



SCIENCE. 



317 



able discussion in the pviblic press, which the 

 committee encouraged in order to bring out 

 all possible objections and as a matter of 

 public education, it requested the president 

 of the Metrological Society to recommend the 

 suggestions in this report to the consideration 

 of the General Time Convention. 



Mr. Allen now took up the subject with 

 renewed energy. By two years of intense dis- 

 cussion and correspondence he was able to 

 enlist the majority of the railroads in support 

 of the Greenwich hour system. The success 

 of the reform rested wholly with him. Presi- 

 dent Barnard distributed widely the various 

 circulars of the American Metrological So- 

 ciety on standard time. The chief signal 

 officer, General William B. Hazen, adopted 

 the idea of the proposed reform with the 

 greatest enthusiasm. The signal office had 

 already found it necessary to instruct some of 

 '"its men in the astronomical methods of de- 

 termining time, and its observers at Dudley 

 Observatory and elsewhere, in addition to 

 their regular meteorological duties, were now 

 regulating local time and even local railroad 

 time when no special astronomers were avail- 

 able for that purpose. In order that the stand- 

 ard of time sent to signal service observers 

 daily should be correct to the nearest second, 

 so that such miscellaneous observations as at- 

 mospheric electricity, auroras, thunder storms, 

 earthquakes, meteors, might not be deficient 

 in accuracy, Professor Prank Waldo suggested 

 the plan and General Hazen authorized what 

 might be called a general clearing house for 

 all the astronomical observatories that wished 

 to cooperate with him. It was arranged that 

 an early morning signal should be telegraphed, 

 say at 11 a.m., and automatically recorded by 

 the chronograph and clock of the Signal Serv- 

 ice at Washington. Among these signals 

 some would always be very accurate because 

 the respective observatories had clear weather 

 and had just made a good time determination 

 during the preceding night, while others would 

 be relatively inaccurate because of cloudy 

 weather at the respective localities where the 

 observatories had been necessarily relying 

 . upon their clock rates for several days. By 

 combining all these into one general mean 



giving appropriate weights it thus became 

 possible to send back to each observatory a 

 statement as to how much its own clock was 

 in error as compared with the mean of all 

 throughout the country. Such a beautiful 

 system as this might with propriety have been 

 inaugurated by any astronomical observatory, 

 but an extensive correspondence had shown 

 that the signal service was the only institution 

 with which all the observatories and astron- 

 omers were disposed to cooperate in this work. 

 That some such plan was, and is now, de- 

 sirable is shown by the fact that a comparison 

 of the standard time signals of the U. S. 

 Naval Observatory, the Harvard College Ob- 

 servatory and the Alleghany Observatory, 

 made each day during several years at New 

 York by the Western Union Telegraph Com- 

 pany, showed large differences, amounting to 

 as much as five seconds in some few cases and 

 frequently exceeding two seconds. 



Unfortunately, however, in 1884, shortly be- 

 fore the clearing-house system was to go into 

 operation, some evil-minded person seems to 

 have induced the secretary of the navy to 

 believe that the Signal Service, in its desire 

 to secure accurate time, was exceeding its own 

 legitimate duties, and an order came from the 

 secretary of war forbidding further action, 

 and even requiring the removal of the elegant 

 clock from its ideal location in the basement 

 of the War Department building. This was 

 the first clock in this country to be so mounted 

 that the air pressure and the air temperature 

 to which it was subjected would remain con- 

 stant. The country was thus deprived, at 

 least temporarily, of the prospect of obtaining 

 the most accurate possible time service for the 

 use of both astronomers and the public, and 

 the U. S. ISTaval Observatory continued its 

 struggle against the use of standard hour 

 meridians. 



Meantime, however, Mr. Allen had induced 

 the railroads to agree that his standard hour 

 system should go into effect on the eighteenth 

 of November, 1883. In order to realize this 

 result it was necessary that the observatories 

 giving time to the railroads and local commu- 

 nities should change to the new system of hour 

 meridians. This change in public clocks and 



