100 Mr. J. Nixon on the Measurevient of the 



tions made with the instrument steadied on the great stone 

 pillar by three huge discs of lead resting within the lower 

 part of its frame. The reversing point of the right-hand 

 level came out on everj' trial very nearly the same, but that 

 of the other level proved more fluctuating; the consequence 

 apparently of some accidental injury done to the case. That 

 no displacement of the adjustment of the zero-line of each 

 index to its arc had occurred during the process of measure- 

 ment was verified by examining through a lens their exact 

 coincidence, and the unbroken appearance of the connecting 

 pellicle of glue. 



The upright wire of the sector was rendered perpendicular 

 in the following manner. At the northern extremity of the 

 plank was set up a white board, carrying in the upper part 

 a projecting nail, from which was freely suspended a short 

 plumb-line. At the southern extremity of the plank stood 

 the round telescope described page 347*, resting within two 

 substantial Ys glued to a thick board. At the middle of the 

 plank was placed the sector, its eye-tube (about a foot distant 

 from the white board) being drawn out until distinct vision 

 was obtained of the plumb-line as seen by the round telescope 

 through the object-glass of the sector. The upright wire 

 being now made parallel to the plumb-line by moving the cy- 

 linder of the sector about within its Ys, the cross levels were 

 adjusted in the usual way f . 



Proof Telescopes i and Method of making their Lines of Colli- 

 mation ■parallel to each other. 



The 30-inch telescope (carrying Fortin's level), which we 

 shall call the square telescope, has been fully described at 

 pages 339 and 345. 



The other, or round telescope just mentioned, was mounted, 

 when one of its cross wires had been rendered truly perpen- 

 dicular, by the 14-inch level-tube glued to one side of the 

 telescope and also to the Ys. Although the glue, which was 

 extremely dilute, had been most sparingly applied, yet more 

 than a week had elapsed before the position of the level-tube 

 relative to the line of coUimation became unquestionably con- 

 stant. 



The square telescope being placed on the plank near its 

 northern end, and a piece of white pasteboard set up a few 

 inches beyond or north of its eye-tube, the round telescope 

 was stationed at the southern end of the plank in a line with 



* All the references are to Phil. Mag. and Annals, N.S. vol x. 

 •)• This method of setting a wire perpendicular is more convenient, yet 

 certainly less accurate than either of those given at pages 341 and 345. 



