880 



SCIENCE. 



[N. S. Vol. XIV. No. 362 



with the prime vertical instrument and 

 vertical circle. These observations, which 

 appear to be of a high degree of excellence, 

 have now lasted eight years, and should be 

 maintained for at least eleven years longer, 

 or through a complete metonic cycle, to 

 furnish a good determination of the con- 

 stant of nutation. An additional com- 

 puter is much needed for the reduction of 

 this work. 



The determination and distribution of 

 accurate time will always be an important 

 function of a government observatory, and 

 requires a careful administration. Not 

 only is the exact time required for rating 

 the chronometers of the Navy, but at pres- 

 ent time signals are distributed by tele- 

 graph from the Naval Observatory through- 

 out the country. The value of accurate 

 time to the people generally is so great that 

 no reasonable effort should be spared to re- 

 duce the inevitable errors from every source 

 as much as possible. Formerly large sums 

 were expended by railroads, cities, factories 

 and makers of clocks and watches in the 

 purchase of time signals, furnished by 

 various observatories. To many observa- 

 tories this was an important source of 

 income. 



In 1883 the astronomers of the country, 

 recognizing the great value of a system of 

 uniform standard time, used all their influ- 

 ence toward securing its adoption, notwith- 

 standing the obvious danger to their private 

 Interests — since under such a system it 

 would be possible for a single institution to 

 furnish time to the whole country. In 

 fact, in 1891 the Naval Observatory as- 

 sumed the task. While the time as now 

 distributed from the Observatory is suf- 

 ficiently accurate for most business pur- 

 poses, since a nonaccumulating error of a 

 few seconds is of little importance in that 

 view, it appears, however, that at consid- 

 erable distances from Washington the 

 errors are often considerably greater than 



under the old system, when the time was 

 communicated from a standard clock not 

 very far away, the errors being often large 

 enough to interfere with the utility of the 

 signals for chronometer rating and scien- 

 tific purposes. The difficulty lies largely in 

 the methods by which the time is trans- 

 mitted by the telegraph, but more in the 

 slightly irregular rate of the standard clock 

 through long intervals of cloudy weather 

 during which star observations cannot be 

 obtained. 



It would seem that by cooperation with 

 other observatories a considerable improve- 

 ment might be made, by arranging matters 

 so that every night, or at least on nights 

 during which the weather is bad at Wash- 

 ington, these observatories should each 

 send a time signal to the Naval Observa- 

 tory, indicating also the interval since the 

 last observation. 



It is difficult to see any way in which 

 the efficiency of the Navy is increased by 

 means of the 26-inch equatorial telescope. 

 If left idle, however, it would be a subject 

 for severe criticism, owing to the large 

 sums of money already expended upon it. 

 Its work should, therefore, be planned 

 wholly in the interests of astronomy. The 

 same principle should be adopted as with 

 the other instruments. Kesearches should 

 be undertaken which have been neglected 

 at other observatories, duplication of work 

 should be carefully avoided, and no investi- 

 gation should be undertaken that can be 

 done equally well with smaller instruments. 

 The following examples of the species of 

 work to be undertaken may be mentioned : 

 A list of double stars which are too difficult 

 to be observed with small instruments, and 

 of which no recent accurate measures have 

 been made, should be prepared with the 

 advice of specialists in this department, 

 and measured with the greatest accuracy ; 

 binaries and suspected binaries should, of 

 course, receive special attention ; neglected 



