Grubb — A New Form, of Position-Finder for Ships' Compasses. 147 



The whole instrument is then turned till the image of the 

 particular landmark, whose bearing it is desired to take, coincides 

 with the pointer ; and then the division on the card which is 

 seen to correspond with the pointer is noted as the bearing of 

 that landmark. 



If the vessel were absolutely steady, and the eye of the observer 

 quite constant in position, such an observation could be made with 

 sufficient accuracy ; but the construction of the instrument is such 

 that it is very difficult to obtain even moderately correct results, 

 except when the above conditions exist. 



The pointer, the compass-card, and the image of the landscape, 

 all of which have to be superposed, are, of course, at different 

 distances ; and no two can be seen distinctly at the same time ; 

 while if the eye be not kept absolutely fixed (a difficult thing in a 

 choppy sea), there will be a considerable amount of parallax and 

 consequent error. 



By employing the same principle of construction as used in my 

 gun-sights and geodetical instruments, the following advantages 

 are obtained: — 



1. The object on landscape or on horizon is seen erect, and not, 

 as in the old form, inverted, which constantly leads to mistakes. 



2. The image of the divided arc is formed (by the action of 

 the collimating lens) in the same plane as that of the landscape ; 

 and therefore there is no parallax, nor any possibility of error 

 arising from the same, nor necessity to keep the eye steady in one 

 fixed place. 



3. In the old instrument two coincidences had to be made : the 

 first between the object and the pointer; the second between the 

 pointer and the division on the compass-card. In the new 

 instrument there is no pointer or index ; but the divisions on the 

 card itself are seen projected on the landscape, and the particular 

 division which falls upon the object is the bearing of that object. 



4. As the collimating lens is made of a focus equal to the 

 semi-diameter of the compass-card, the degree-divisions on the 

 card correspond to degrees on the horizon ; consequently it is not 

 necessary (as in the old form) to set the instrument on the binnacle 

 with the pointer exactly coinciding with the object (another 

 difficult operation in a rough sea). If the object be anywhere in 



