CHAP. IV,] 



MAKING A MAIN STATION 



85 



should be secured, first, because later, when they fall into 

 shadow, they may be wholly invisible. 



In most instances a sketch will be also necessary, on which 

 the angles to conspicuous objects, tangents, etc., "will then be 

 recorded, instead of in the book. 



All angles should be read twice, in order to prevent mis- Repeti- 

 takes ; but to ensure accuracy when required, the angles must ^ngigg 

 be repeated on different parts of the circular arch, for the follow- 

 ing reasons : 



A theodolite, however well turned out, is seldom exactly 

 centred, hence arises error, as, no matter how uniform the 

 graduation of the circular arc may be, a slight deviation of 

 the axis from true centring wiU give a difference of reading for 

 an angle on different parts of the arc. 



The mean of the readings of the two verniers is supposed to 

 correct errors of centring, but for remarks on this, see 

 "Theodolite." 



The reading itself of an angle can never be considered as 

 perfectly correct. Slight parallax frequently exists, especially 

 when an instrument has been some time at work, and is 

 getting worn. In small theodolites the marking of vernier 

 and arc at any given angle will often not coincide exactly, and 

 judgment may assign the wrong reading. 



By multiplying readings, then, a mean will be obtained 

 which will to a great extent eliminate these errors, and this 

 must always be done in observing main angles. 



Excepting for main angles, forms ruled in the angle book 

 are unnecessary, and m this case the form is simple, consisting 

 of columns to keep the figures separate, as under. 



h July 4th, 1881, at Pagoda A, Theod. 77. 



