On the Use of the Pocket Box-Sextant. 



59 



Lat. deduced from ob-i 

 servations made with 

 an observatory instru-r 51 20 5.6 . 

 ment,and by measure- 

 ment from observatoryj 



Lat. too little by sextant 25.246 



Tlius we see that a mean of 24 observations give the latitude 

 within about 25" of the truth : with fewer and more imperfect 

 observations, a traveller might give us a tolerable account of the 

 latitude in which he travelled, and we all know how frequently 

 this is uncertain to a whole degree or two. 



With respect to longitude, it cannot be expected to be deter- 

 mined by lunar observations, by a single person and a single 

 instrument of any kind, so as to become a criterion of the accuracy 

 of that instrument ; but, as the time has been always obtained 

 within a few seconds, and sometimes agreeing in two observations 

 to a single second, great hopes may be entertained of the possi- 

 bility of procuring a near approximation to that also, by the use of 

 two of these sextants, especially when reading to 30", and if fur- 

 nished with small telescopes, without which they are certainly not 

 so much to be depended upon. 



The follownig independent results were obtained from separate 

 observations, and may serve to corroborate what has been above 

 advanced : — 



Oct. I. 

 Q 1st obser. O u. 1. \''- „% 



watch fast . .3 



7 2d 14 24 



March 30, 1823. 

 1st obs. O 1 



fast 

 2d .... 13 52.20 

 3d .... 13 50.32 

 4lh . . . .13 47.36 



Sept. 1,1823. 



1st obser. O 1. 1. ) 

 watch fast . 3 



2d 



Sept, 20, a aq. watch -j 



fast • • •> 



1st observation J 



2d 



3d 



Meant of 3 



9 



8 51.5 



9 10.5 



9 0.66 



•.} 



13 47.52 



Mean of 4 



13 49.35 



