310 



HYDROGRAPHICAL SURVEYING [chap. xiii. 



The rule for giving A and B their proper algebraic signs is 

 as follows : 



Signs of 

 Equatioa. 



At 

 Superior 

 Transit. 



A is 



Bis 



+ if declination is decreasing and of same 



name. 

 + if declination is increasing and of different 



name. 



- if otherwise. 



+ if declination is "i 



increasing I When elapsed time is 



- if declination is j less than 12 hours. 



decreasing j 

 Reversed when elapsed time is greater than 

 12 hours. 



Noon 

 Com- 

 parison 

 not to be 

 used. 



At 

 Inferior 

 Transit. 



Change of 

 Declina- 

 tion in 

 Inferior 

 Transit. 



Calculat- 

 ing a 

 Mean 

 Com- 

 parison 

 for Hack- 

 Watch. 



The above rule for the algebraic signs of A and B may be 

 stated more briefly thus : 



A is 4- from summer to winter solstice, and vice versa. 

 B is + the equinoxes and the solstices, and vice versa. 



A is reversed from what it would be at superior transit. 

 B is the same as at superior transit. 



In working with inferior transit, whereby we find the Error 

 at midnight, there is no difference in the rule, except that in 

 calculating the change during the half-elapsed time, we use the 

 variation of declination found by interpolation for the Green- 

 wich time of local midnight. 



The next step is to calculate, from the comparisons taken 

 with the standard before and after sights, a mean comparison 

 to apply to the Error of Avatch found above, to arrive at the 

 Error of the standard. 



To do this, we take any sight, and by interpolation calcu- 

 late the comparison at the a.m., and also the p.m. time corre- 

 sponding. The mean of these two will give the comparison at 

 noon. This should, if the w-atch has been going mcU, corre- 

 spond very closely with the comparison actually taken at noon, 

 and it will be satisfactory if it does so. If it does not, we 

 cannot help it ; but we shall know that the Error of the standard 

 will be slightly incorrect from a jump in the watch, and shall 

 be prepared to give the result a smaller value in consequence, 

 in event of discrepancies with others. 



The mean comparison, as found above, must always be used, 

 not the comparison taken at noon, which is done solely to ascer- 

 tain how the watch has been going. 



