OF THE OBSERVATORY OF CAMBRIDGE BY GALVANIC SIGNALS. 511 



Hence the interval apparently occupied by the transmission of the galvanic action is 

 by this calculation 0*,0195. I think the former result to be rather more trustworthy. 



The results of the method (B) may be deduced from those of the method (A), by merely 

 applying the differences of the errors of the transit-clocks at the same epochs as given by the 

 two methods, since the rates may be assumed to be the same as determined by either method. 

 Now on May 17, the Greenwich transit-clock was ! ,10 slower, and the Cambridge transit- 

 clock S ,05 slower, by method (B) than by method (A). Hence - 0*,05 is to be applied to 

 the longitude found by the latter method to obtain that given by the former. On May 18, 

 the Greenwich transit-clock was s ,06 slower, and the Cambridge transit-clock 0%l6 slower, 

 by method (B) than by method (A), and the correction is consequently +0%10. The final 

 results are, therefore, as follows : — 



East Longitude of the Cambridge Observatory. 



Method (A). 



». 

 May 17, by 146 signals 22,922 



18, by 135 signals 23,038 



Mean 23,015 



Method (B) ; 



May 17, by 146 signals 22,942 



18, by 135 signals 23,138 



Mean 23,040 



Mean of the whole 23,027 



Shortly after the experiment was made I calculated the longitude provisionally, and 

 communicated the results to the Astronomer Royal, stating at the same time that they would 

 probably be modified by calculations I proposed to make respecting the effect of the forms 

 of the pivots of the Cambridge Transit. The difference between these first results (which 

 are published in the Monthly Notices of the Royal Astronomical Society, Vol. xm. No. 8, 

 p. 251) and those obtained above, is owing partly to slight mistakes discovered in the former 

 calculation, and partly to the circumstance that it corrected in an imperfect manner for the 

 forms of the pivots, whilst the present calculation supposes the transit-pivots to be 

 cylindrical and equal. Having recently gone through a complete investigation of the 

 corrections required to be applied for inequalities of the pivots, according to the method 

 which I have descrided in the Memoirs of the Royal Astronomical Society (Vol. xix. p. 103), 

 I will now state the effect of these corrections on the determination of the longitude, and 

 then add the evidence on which the necessity for such corrections rests. 



The forms of the pivots being taken into account, the following are the errors of Hardy: — 



Method (A). 



A. m. m. i. 



May 17, at 9.49 sidereal time, Hardy slow 1.20,328 



18, at 14. 9 1.22,243 



Vol. IX. Part IV. 66 



