CHAP. XTT.] LATITUDE BY SUN 289 



In observing stars under the pole, we must not forget that stars 

 the reduction will be subtractive from the observed altitude. p°fe®^ 



The larger planets are not good for observation, as they are Planets, 

 so much bigger and brighter than the point which a star 

 shows. On occasions, however, they must be used. The R.A. 

 and Dec. must be calculated exactly. 



In using stars from the Greenwich or Cape Catalogues, it is Calculat- 

 necessary to calculate the apparent place of the star for the Apparent 

 day. The method of doing tliis is given in the " Nautical Places of 

 Almanac," in the explanation under the head of " Stars," fpom^ 

 where examples of a north and of a south star are shown Cata- 

 worked out. Care must be taken to give the proper signs °^®^" 

 + or — to each logarithm. 



Next to the observation of stars in pairs, the circum-meridian Latitude 



observation of the sun in the artificial horizon is the most ^^ ^^^' 



cum- 



correct and simple method we have of obtaining latitude ; but Meridian 

 it is evident that we cannot use it when the altitude exceeds f/gun'^f ^ 

 65°, as a sextant will not measure the double angle. We must, 

 in the case of the sun, be doubly careful in correcting the 

 refraction if we wish to get as near the truth as possible. 

 There is nothing to be gained by observing both limbs of the 

 sun, as the motion in altitude will be so small that it will not 

 matter whether the images are opening or closing. 



The roof of the horizon should be reversed at about noon, 

 and the sights worked out as two sets, roof one way and roof 

 the other. 



However careful we may be, we shall not expect our latitude 

 by the sun only to be exact, and in many cases where we are 

 going to be satisfied with this observation it will not matter 

 if the latitude be a quarter of a mile or so in error, and the 

 reversal of the roof may often be dispensed with. 



If, however, we know our centring error, and can depend 

 upon the sextant, we can by its application get a vastly im- 

 proved result, and none of these precautions should be omitted. 



An observation of the sun cannot be meaned with an observa- Sun and 

 tion of a star the other side of the zenith, as all refraction not^g'^*" 

 errors, as well as errors introduced into the instrument by the Paired, 

 heat of the sun, will be entirely different. 



Circum-meridian observations of the sun are worked out in 

 precisely the same manner as those of a star, the only dif- 



19 



