438 MEMOIRS OF THE AMERICAN ACADEMY. 



the charge of Mr. George W. Dean, of Fall River, Mass., assisted by Mr. F. Blake, 

 experienced officers in the Coast Survey Service. Mr. Dean acknowledges, in his 

 report, the obligations of the Survey to Mr. Thomas Andrews, General Superintendent 

 of the French cable at Brest, for his active and cheerful co-operation in the objects 

 of the expedition, and also to Mr. J. D. H. Dickson for valuable aid in arranging the 

 batteries and other instruments in the cable office. 



The U. S. Coast Survey station at Brest is situated near the southeastern part of 

 the grounds attached to the Mablissement des Pupilles de la Marine, and 126.44 

 metres west of the flag-staff on the tower of the St. Louis Church, which served as 



one of the principal points in the primary triangulation of France. Mr. Dean con 

 nected this flag-staff with his station by a triangle, with a measured base of 117.33 



metres, extending due north from his own position. The Coast Survey Station was 



distant about one quarter of a mile from the office of the Cable Telegraph, and was 



temporarily connected with it by two wires. Its latitude is 48° 23' 21 // .4 N. 



For the determination of the local time at Brest, U. S. C. S. Transit No. 4, made by 

 Troughton and Simms of London, in 1848, was used. Its principal focus is forty-six 

 inches in length, and its aperture two inches and three quarters of an inch. The 

 power employed was ninety-five. It was noticed that some of the threads, particu- 

 larly the middle one (D 3 ), were disturbed by the hygrometric changes of the atmos- 



lere. The level marked (B) was used for ascertaining the inclination of the 

 mechanical axis of the instrument. The arc-value of one division of the scale was 

 found in 1868 to be 0.99 of a second of arc, or 0.066 of a second of time. It was known, 

 by observations made at Salt Lake City in 1869, that the lamp-end pivot was the 

 largest, and required a correction amounting to S .013. The thread-intervals were 

 subjected to a new determination, which will be given in another part of this paper. 



A Sidereal Chronometer, made by Charles Frodsham of London, and numbered 

 3451, was loaned to the Coast Survey Service by Professor Joseph Winlock, Director 

 of the Harvard College Observatory. This instrument is provided with a break- 

 circuit, invented by Mr. Frodsham in 1868, so as to record the alternate seconds upon 

 a chronograph-sheet ; but at the beginning of each minute two successive seconds are 

 registered. Mr. Dean observes that this instrument, though its rate was not so uni- 

 form at Brest as at the Observatory, has on the whole performed satisfactorily. The 

 Chronograph employed (which was U. S. C. S. No. 1) worked well. 



The astronomical station of the U. S. Coast Survey at Duxbury, Mass., was under 

 the charge of Mr. Edward Goodfellow. It is situated on ground belonging to Mr. 

 William Paulding, at a distance of 881 feet E. N. E. from the office of the Cable Tele- 



