CHAP XXL] ASTRONOMICAL OBSERVATIONS 497 



In order to obtain any degree of precision it is necessary to 

 observe every precaution, amongst which may be mentioned: 



Each observer to use the same instrument throughout, 

 whose errors are satisfactorily ascertained from time to time. 



The observer's eye from a protected position to be always 

 at the same height above the sea, and this as low as possible 

 as will enable a clear definition of the horizon with due regard 

 to the then state of the sea and motion of the ship. 



Before commencing observations it is often very necessary 

 to scan that part of the horizon upon which we are dependent 

 carefully with a pair of binoculars. 



This is more especially the case when passing clouds or 

 showers give way to bursts of sunshine, or in calm weather when 

 haze or the smoke of passing vessels lie in a horizontal strata. 

 An inspection will often reveal a false horizon due to these 

 causes, which otherwise would not have been detected, and 

 thus necessitate waiting until better conditions prevail, al- 

 though the Ukelihood of the heavenly body remaining unob- 

 scured appearing uncertain would form an inducement not to 

 do so. 



SECTION XVII. 



CHECKING THE POSITION OF THE BEACONS AND THE SCALE 

 SUBSEQUENT TO THE ORIGINAL TEIANGULATION. 



It should be an invariable rule, throughout the progress of a 

 survey dependent ujDon beacons, for all the vessels engaged to 

 check the positions of the beacons at an early opportunity each 

 day, on commencing work, or if remaining at anchor, when an 

 all-round fix taken to all the beacons that are then in sight should 

 be tested. Whilst sounding is in progress, check angles 

 when fixing, the occurrence of transits, compass bearings ; 

 simultaneous fixes, using different objects, will serve for this 

 without causing any undue delay. 



It is not always an easy matter to determine at once how 

 many, and which, of the beacons at any time no longer occupy 

 their original positions when a discrepancy has been revealed. 

 As soon as the " erring beacons " can be identified, and their 

 new positions ascertained, they should be treated with caution 

 as " doubtful quantities," likely to move again, probably owing 



32 



