Notices respeclvig Neio Looks. 289 



ledge of the temperature is necessary) was indicated by both methods 

 with a precision equal to that with which the absolute lengths of the 

 bars can be determined. In the case, indeed, of the long steel chains 

 formerly used in the Ordnance Survey, the uncertainty respecting 

 the temperature of the different parts of the measuring chain forms 

 a serious objection; but with respect to simple bars of a moderate 

 length, the practical difficulty of determining the expansion at the 

 moment of the measure does not seem insuperable. 



The following is the description of the measuring bars, omitting 

 the references to the plates : — 



" The compensation bar consists of two bars, of brass and iron, 

 10 feet I'd inch long, 0'5 inch broad, placed l'r25 inch apart, sup- 

 ported on brass rollers at one-fourth and three-fourths of their length, 

 and firmly fixed together at their centres by transverse steel cylin- 

 ders rs inch in diameter, and being free to expand from, or contract 

 towards, their centres independently of each other. At the extremity 

 of, and at right angles to, each of these bars is a flat steel tongue 

 6-2 inches long, 1-1 inch broad, and 0-25 inch thick; projecting 

 3"25 inches on the side of the iron bar, and moving freely on conical 

 brass pivots, riveted into the brass and iron bars, each axis being 

 perpendicular to the surface of the tongue, allowing it to be inclined 

 at slightly different angles to these bars according to their expansion 

 from, or contraction to, their centres. The centres of the two axes 

 are at O'o inch and 2 '3 inches from the end of the tongue next the 

 brass bar. On the tongue, and flush with its upper surface near the 

 (projecting) extremity, is inserted a silver pin, with a dot marked on 

 it, as the compensation point." — P. 10. 



The number of such bars used in the measurement was six. 



Besides the compensation of the expansion of the measuring bars, 

 and the consequent avoidance of all reductions to the normal tem- 

 perature, another object, which was considered of importance, was 

 aimed at in the new apparatus. " To avoid the possibility of acci- 

 dental displacements incident to the ordinary mode of measuring base 

 lines by the contacts of the ends of rods, or by the adjustments 

 of the coincidences of lines, it was deemed advisable that the several 

 ])arts of the apparatus should be combined by visual contacts." 

 Accordingly an interval of about six inches was left between the 

 measuring bars, when placed in the line of the base, for the accurate 

 measurement of wliich a microscopic apparatus, also on the principle 

 of compensation, was contrived. It is described as follows : — 



" The compensation microscope consists of three microscopes, 

 placed 3 inches from centre to centre, connected by two bars of 

 brass and iron, 7 inches long, O'G inch broad, and 0'375 inch thick, 

 2'.5 inches apart, firmly secured together by means of a brass collar 

 and cylinder, forming part of the tube of the centre or telescopic 

 microscojic ; the two bars carrying with them the outer microscopes, 

 of two inches focal distance, being free to expand from, and contract 

 towards tlie central microscope, independently of each other, and 

 thereby forming with it small angles of inclination similar to tlic 

 steel tongues of tlic compensation bars. The compensated i)oint of 

 Phil. Mag. S. ."?. Vol. 32. No. 21.5. /Jpril IBIS. U 



