Proceedings of the Royal Institution, 387 



The mean of the best measurements may be taken at about -g^. 

 The amount deduced by Laplace in a manner totally different, by 

 separating and estimating those inequalities in the moon's motion 

 due to the excentricity of the earth, is y^. 



The first and perhaps the most dilRcult of these operations con- 

 li^ in the measurement of a base line, on the accuracy of which 

 depends that of every other distance, and finally, of the whole meri- 

 dional arc. The improvement in the apparatus used for this purpose 

 has been great and progressive, from a simple wooden rod laid on 

 the ground without supports, without correction for moisture, tem- 

 perature, or curvature, to the beautiful and delicate apparatus in- 

 vented by Lieut-Colonel Colby, and executed with great ability by 

 Messrs. Troughton and Simms. This apparatus consists of six bars each, 

 10 feet in length, and constructed on a principle of compensation, so 

 as to maintain one invariable length during every variation of temper- 

 ature. This is effected with the aid of two different metals, viz. 

 brass and iron, and according to a simple, though very efficient ar- 

 rangement. 



The bars, when placed in line, are not in contact, but an interval 

 of six inches intervenes between each, which intervening space is 

 measured by two microscopes, arranged in a similar manner to that 

 of the bars, the foci being, in this case, the two points whose distance 

 from each other remains invariably the same. The six bars with 

 their microscopes, forming a length of 63 feet, were placed for the 

 most part on one level, and the next series rose or fell according to 

 the nature of the ground. This rise or descent took place in vertical 

 lines, and was effected by an optical contrivance, by means of which 

 it could be accomplished with nearly equal facihty for any distance, 

 between 4 inches and 8 feet. 5 



. The means adopted for rendering the principle of compensation 

 perfect, by equalising the rate at which the metals change their tem- 

 perature ; the various mechanical adjustments and contrivances re- 

 quired to render the principle just described of easy and certain appli- 

 cation, would require, in order to be properly understood, a more 

 detailed description than the limits of the present notice will permit. 

 But we may add, that this method was in principle the same as would 

 be adopted for determining the length of a standard scale, and hence 



