CHAP. XXL] DISTANCE BETWEEN POSITIONS 509 



mark on the bowsprit end, etc., all of which are suitable for 

 subsequent use when determining the distance of the observer 

 from the ship by means of the horizontal angle subtended be- 

 tween them. These objects should therefore also be intersected 

 with the theodolites; but as the former are liable to removal, 

 alteration, or damage, we shall depend upon the deck marks, 

 from which any desired extension can be made at any future time 

 by actual measurement on board. 



ReUable results are equally obtainable under favourable 

 conditions, such as a calm and no tidal stream, when the ship 

 is at anchor and the observers stationed in readiness at suitable 

 positions on shore, not too far distant from the ship. 



It is generally impossible to make any du-ect measurement 

 on board between two positions situated near the ends of the 

 ship, whether they are selected out of the central line or not, 

 owing to the numerous objects that generally intervene. 



If it is desired to determine this distance without resort to 

 some measurement made on shore, a reUable result can be ob- 

 tained by selecting successive intervisible theodoUte positions 

 at suitable intervals and at about the same relative height, 

 between which rehable measurements can be made either along 

 the intervening straight deck or by suspending wire, etc. 



The computation of the horizontal distance between the 

 selected extreme points, which need not necessarily be inter- 

 visible, is the solution of the connected traverse made between 

 them. 



Before finally removing the instruments it is as well to make 

 sure that all the necessary data for this have been obtained, and 

 that all the steps are complete. 



When great accuracy is required, or should the measured 

 distances depart much from the horizontal, the distances must 

 be referred to the horizontal plan ; readings taken on the vertical 

 arc of the theodolite will suffice for this, and it is therefore 

 necessary to select a time when the ship is practically motionless. 



In all cases the draught of water of the vessel should be 

 carefuUy recorded, and the vessel as nearly upright as possible. 



