QUANTITY OF HORIZONTAL REFRACTION. ^<J 



On account of the difficulty attending fome of the adjufl-Objeflionsto 

 ments for the back obfervation, he rejeas that method for the back obfer- 



r ^ 1 \ II vation or angles, 

 taking angles in general, with much reaion ; but he has thereby 



overlooked a means of determining the dip, which I am inclined 

 to think might be employed with advantage in all latitudes, 

 without any occafion to hurry the mofl: inexperienced or cau- 

 tious obferver. 



By the back obfervation, the whole vertical angle between but the method 

 any two oppotite points of the horizon may be meafured ^ meafuring the""^ 

 once, either before or after taking an altitude. Half the ex- arc between the 



cefs of this angle above 180°, (hould of courfe be the dip re- °PP°'"^t^ h°"- 



o ' ^ tons. 



quired. 



But, if it be doubtful whether the inftrument is duly adjufled. Error of the 



a fecond obfervation becomes neceflary. The inftrument muft ^ I , ^ 



J reverfing the 



be reverfed, and, if the apparent deficiency of the oppofite whole inftru- 

 angle from 1 ^0® be not equal to the excefs before obtained, "^ent, 

 the index error may then be corre6led accordingly ; and, fince 

 the want of adjnftment, either of the glafles at right angles to 

 the plane of the inftrument, or of the line of fight parallel to it, 

 will affeft both the larger and fmaller angle very nearly in an 

 equal degree, the ^ part of their difference will be extremely 

 near the truth, and the errors arifing from want of thofe ad- 

 juftments may with fafety be negledled. 



This method of correding the index error for the back obfer- as firft fuggcftei 

 vation at fea, was many years fince recommended by Mr. ^V Ludlam. 

 Ludlam ; * yet I do not find that it has been noticed by fubfe- ' 



quent writers on that lubje<5t, or fuggefted by any one for 

 determining the dip ; but I can difcover no reafon for which it 

 could be reje6led as fallacious, and I ftiould hope that in prac- 

 tice it would be found convenient, fince in theory it appears to 

 be effeftual. 



The moft obvious objedion to this, as well as to Mr. Hud- Whether refrac- 

 dart's method, is the poffibility that the refradion may be in ^'°" ""^^ ^''.f'^^ 



,.rr r • - ferent on dif- 



fome mealure different in oppofite pomts ot the horizon at the ferent bearings 

 fame time. When land is at no great diftance, fuch an in- ^P ^^^ ^^^^ 

 equality may be found to occur ; but, upon the furface of the 

 ocean in general, any partial variations of temperature can 

 rarely be fuppofed to exift ; and it is probable, that under any 

 circumflances, the difference will not bear any confiderable 



* Dire(5lions for the ufe of Hadley's quadrant, 1771, § 82, p. 5S. 



proportion 



