[203]- 



the fum of the intervals of her A. R. from that of the ftar is 

 to be ufed in this calculation inftead of the difference of thefc 

 intervals. 



When the meridians are even a moderate diftance from each 

 other, the negledl of the equation for the difference of the 

 obferved and true interval will have a very confiderable in- 

 fluence on the refult of the obfervations ; when the moon moves 

 at her quickefl rate, and the meridians are only 10' afunder, 

 - the corredtion amounts to 5" of time on the obferved interval, 

 it will of courfe in all cafes require to be very flridly attend- 

 ed to. To facilitate therefore this part of the operation, I have 

 calculated a fhort table, intitled " A table of corredions, &c." for 

 which the corredion of the interval for any lunar rate, and for 

 any longitude not exceeding 10^ may be taken out with fuf- 

 ficient accuracy almoft at fight. The corredion in this table 

 is given in time to feconds, tenths, &c. that it may be diredly 

 applied to the difference of the A. R's. as found immediately 

 by the tranfit inl^rument. The difference thus correded is to 

 be changed into parts of a degree, to prepare it for the ana- 

 logy from whence the longitude is to be finally deduced. 



We have hitherto confidered the moon's rate in her orbit to 

 be equable, but as this is feldom in reality the cafe, this me- 

 thod requires that when her rate is unequal, the fum of her 

 A. R, gained in 12H. fliould be equated as nearly as poflible, 



C c 2 according 



