ON ASTRONOMY. 101 



" A number of stars, culminating near the zenith of the two stations, 

 are selected by the observers. As they appear first upon the eastern 

 meridian their transit is recorded by the observer striking the finger- 

 key upon the chronograpliic registers at both stations. Alter an 

 interval of time equivalent to the difference of longitude between the 

 two places, which is measured by the clock, the same stars appear on 

 the western meridian, and the observer at that station records this 

 transit precisely as the other had done, and the dilference of the two 

 records of time is the measure of the difference of longitude. 



" It will be observed that these records have been obtained at both 

 stations, and a little reflection will show that if there be any sensible 

 interval of time consumed in the transmission of signals, the differ- 

 ence of longitude obtained from the record at the eastern station will 

 be too great by that interval, and that at the western station will be 

 too small by the same amount. The mean result will give the longi- 

 tude tree from this error, and the difference of measure the time of 

 transmission of the signals through the whole circuit. 



"Ten stars are generally exchanged with the eastern clock in the 

 circuit, and, after the first five, the transit instrument is reversed so 

 as to eliminate any residual error in the correction for collimation. 

 The western clock is next put on and the same operation repeated 

 ■with ten other zenith stars. Not only is the result improved by the 

 accumulation of individual results, but the advantage is gained that 

 the interval is measured by another clock, and the time of transmis- 

 sion eliminated in the inverse order of effects. The transits of the 

 stars are generally recorded over fifteen wires. After the exchange 

 of signals by the second clock is completed, local observations for 

 instrumental corrections are again made, which conclude the night's 

 work. These operations are repeated on, at least, three different 

 nights, after which the observers and instruments exchange places, so 

 as to eliminate the possible errors arising from causes connected with 

 their individual peculiarities. 



" By the perfect and admirable method just sketched we are able to 

 measure arcs of longitude with the same degree of accuracy with which 

 arcs of latitude have heretofore been ascertained, and a new element 

 has thus been introduced into geodesy. Since the general introduc- 

 tion of the electric telegraph and the development of the American 

 method of longitudes it has been applied to many of tlie oldt^r Euro- 

 pean geodetic surveys ; and, in general, a very full acknowledgment 

 has been made of their indebtedness to American science by the emi- 

 nent geometers having charge of such works. 



"It was one of the earliest discoveries resulting from the telegraphic 

 determinations of longitude that the time transmission of signals 

 between stations several hundred miles apart is quite sensible, and 

 that it appears, in a great degree, to depend on the distance. On the 

 telegraph lines used in this country, where large iron wire is employed, 

 and the circuit is completed through the earth, the rate of transmis- 

 sion has been found to be from 11,000 to 20,000 miles per second. 

 Whether this actually was the velocity of the galvanic current, or 

 whether it rather measures the time during which the current must 



