now measuring on the Continent. 181 



cessively and independently determined ; when the sum of the 

 several portions is the ampHtude due to the total arc. The 

 disadvantage under which a result so obtained labours is, that 

 it is necessarily chargeable with the accumulated sum of the 

 errors, which may have occurred in the determination of the 

 time, at each extremity of each of the portions into which the 

 arc is divided: in which accumulation, however, the errors 

 that have contrary signs will liave mutually destroyed one 

 another, proportionally to their respective amounts. As the 

 same method was adopted in all the portions into which the 

 arc between Marennes and Fiume was divided, differing only 

 in the heights and in the distances apart of the stations, and 

 occasionally in the number of signals employed, we shall limit 

 ourselves to a notice of the stations, and of the periods at 

 which the several determinations were accomplished. 



By means of signals made at La Ferlanderie in October, 

 1823, the difference between the meridians of the steeple of 

 Marennes, and a temporary observatory established at the 

 mill of St. Preuil, was ascertained by M. Nicollet, and Co- 

 lonel Brousseaud, as follows : — 



October 6th, mean of 10 signals 3' 48'',831 



15th, 10 signals 3' 48",969 



16th, 10 signals 3'49',207 



17th, 16 signals 3 49,030 



Mean of the 4 series, 46 signals 3' 49",012 



Reduction to the steeple of Marennes . . . . — 00 ',022 



Difference of meridians between the Observatory 



of St. Preuil and the steeple of Marennes. . . 3' 48'', 99 



In the month preceding, namely, in September, 1823, the 

 difference between the meridians of the observatory at St. 

 Preuil and a temporary observatory established at La Jonchere, 

 was ascertained by the same observers, by signals made on the 

 Puy Cogneux. In this instance, from accidental circum- 

 stances, the signals were visible at both stations on a single 

 night only, that of the 10th of September* ; and the arc between 



* In selecting the summit of the Puy Cogneux as a station for the 

 signals, Colonel Brousseaud had assured himself, by personal observa^ 

 tion from the summit of the mountain, that the mill of St. Preuil was 

 visible from it, and had concluded, conseciuently, that witliout doubt th^' 



