396 Intelligence and Miscellaneous Articles. 



plishing the double of the distance between the two stations. This 

 distance, carefully determined, was found to be 10310 metres, with 

 a probable error of less than 10 metres in excess or defect — that is 

 to say, an approximation to witliin one thousandth part. I effected 

 this measurement myself, with the aid of a little triangulation. I 

 took advantage of the circumstance that the three salients of bas- 

 tions No. 1, No. 2, and No. 5 of the fortifications of Mont Yalerien 

 could be seen from the belvedere of the Ecole Polyfcechnique. Ta- 

 king, with a good azimuth-circle, the angles subtended by those 

 three points, I was able, by the calculation of the proper segments, 

 to obtain the data necessary for ascertaining the distance between 

 the two stations. The dimensions of the fortress having been ac- 

 curately measured by the engineers, I obtained from the Depot des 

 Fortifications the distances between the necessary points, which 

 enabled me to dispense with measuring a base. 



I found, besides, at the Prefecture of the Seine, two determina- 

 tions of the distance from certain points on Mont Yalerien to the 

 Pantheon (one derived from the operations for the Register, the 

 other from those of the Commission for the Plan of Paris). The 

 mean of the three closely agreeing values thus obtained gave the 

 number above adopted. If a subsequent geodesic operation should 

 furnish more exactly the distance between the two stations, it will 

 be very easy to calculate the correction to be applied to my results. 



The following Table gives the result of the definitive calculations ; 

 the values of the velocity of light, expressed in kilometres per se- 

 cond, are classed according to the order of the occultations of the 

 return-fight which furnished them. 



Order 1. 



Order 2. 



Order 3. 



Order 4. 



Order 5. 



Order 6. 



Order 7- 



302600 



297300 



298500 



298800 



297500 



300400 



(17) 



(236) 



(376) 



(480) 



(91) 



(27) 



The numbers in parentheses express the relative weight of the 

 corresponding values. They were formed by dividing by 10 the 

 product of the number of observations into 2n — 1 (n being the order 

 of the occultation) and the coefficient 1, 2, 3, or 4, according as the 

 remark in the experiment-book was " pretty good," " good," " very 

 good," or " excellent," according to the state of the atmosphere. 



The compound mean gives 298400. Multiplying this by the 

 index of refraction of the air, 1*0003, we obtain the number 298500 

 kilometres per second as the value of the velocity of light in vacuo 

 deduced from the totality of my observations. I reckon that this 

 number approaches within ^-q. 



The accordance of this result with Foucault's is not uninterest- 

 ing; and it should be remarked that Poucault's experiments re- 

 quired verification, not only because the details of the observations 

 and procedure have not been published, and therefore have under- 

 gone no discussion, but also because the rotating-mirror method is 

 exposed to grave objections (into the statement of which I shall not 

 now enter), while M. Pizeau's method is free from those objections. 



