1872.] 



for the Indian Trigonometrical Survey. 



325 



being about 6*2 lbs., and of course T l - instead of \ of the whole. An 

 external deep groove, filled with oil, into which a thin circular rib dips, 

 effectually excludes dust from the relieving apparatus and from the vertical 

 axis. 



The proper amount of relief has been a question much discussed. Sir 

 George Everest, in his work above quoted *, fixes it, somewhat arbitrarily, 

 at | of the whole moving mass. This cannot be accepted as a general rule, 

 because, if the mass were very great, £ of its amount might be sufficient to 

 cause a most destructively abrasive action . 



My method of adjusting the relief is as follows : — I assume that, for the 

 purposes of horizontal angles only, the azimuthal motion cannot possibly 

 be too free. I therefore first make the relieving force nearly equal the 

 weight of the moving mass. Then I try the levelling of the instrument. 

 "With such excessive relief this is sure to be unsteady. I then cautiously 

 diminish the relieving force until I obtain steady levelling. When I have 

 done this, I know that I have given all the relief which is possible in the 

 existing construction, consistently with the general purposes for which the 

 instrument is required. In the present instrument the result stands 

 thus : — 



Weight of moving mass 284 lbs. f 



Relief (40 springs X 6-2 lbs.) 248 „ 



which leaves nearly g of the moving mass active, instead of as prescribed 

 by Sir G. Everest. 



Horizontal Tangent- screw. — This, although apparently a mere subor- 

 dinate agent, is really one of great importance and presenting some diffi- 

 culties. There is always " loss of motion " in every screw unprovided with 

 special means to prevent it. Loss of motion increases the difficulty of ob- 

 servation and also causes insecurity ; for the tangent-screw is the link which 

 connects the moving with the fixed parts, and the length of this link, once 

 adjusted iu observation, should be unalterable. Several methods of obvia- 

 ting loss of motion have been tried : and^though many of them are suitable 

 to small instruments, none have been quite satisfactory with large ones. The 

 matter attracted great attention when I was in the Indian-Survey Depart- 

 ment, and I was much engaged in studying it. The general conclusion I 

 came to was that, however a spring might be applied, its tension ought to 

 be constant and invariable. Hitherto this had not been the case, an ordi- 

 nary external antagonizing spiral spring having been used, the compression 

 of which was continually being varied by the action of the tangent-screw. 

 It was found that after the observation had been made the spring continued 

 to act, which was fatal to the observation. And the evil was brought home 

 to this agent by the fact that in a particular instrument the springs attached 



* Page CTiii. 



t Since this adjustment was effected, the weight of the moving mass ha9 been 

 increased by additions to 299 lbs., and the springs will have to be readjusted ; but the 

 absolute weight left active will probably remain nearly as at present. 



