APPLICATION OF THE ELECTRIC TELEGEAPH. 335 



difference of longitude, one hundred and twenty-four ob- 

 servations were made upon thirty-five zenith stars; 

 eighteen observations for collimation and nine for 

 equatorial intervals were made upon three circumpolar 

 stars. In connection with these operations, one hundred 

 and twenty-two observations were made upon twelve 

 stars for local time. The instrument employed was a 

 forty-three inch transit instrument ; the diaphragm con- 

 sisted of twenty-five wires arranged in groups of five. 

 The result of these observations shows Petersburg to be 

 1m. 35'603s. west of Washington, with a probable error 

 of +_ 0*009s. The probable error of a single set of ob- 

 servations of a star over fifteen wires is _+_ OOSls. The 

 residual probable error of a single night's work, with 

 the transits of fifteen stars telegraphed from and received 

 at both stations, not accounted for by the error of tap- 

 ping and receiving, is but O004s., or may be considered 

 insensible, so that it is unnecessary to multiply the num- 

 ber of nights of observation. 



During the winter of 1852 and '53, a second determi- 

 nation of the difference of longitude between Charleston, 

 South Carolina, and the Seaton station in Washington 

 was attempted under the direction of Dr. B. A. Gould 

 From the middle of December until the middle of Feb- 

 ruary, Dr. Grould remained in Charleston, observing on 

 every fair night circumpolar stars with reversals, and 

 both zenith and equatorial stars chrpnographically, for 

 the determination of time and instrumental corrections. 

 Whenever the sky was unclouded both at Washington 



